JP2017125840A - Coloration reaction detection system, coloration reaction detection method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a coloration reaction detection system for easily obtaining inspection results with respect to multiple items by discriminating patterns of the multiple items of an analysis kit without using a special and expensive inspection device and from images captured by a simple imaging apparatus such as a general-purpose digital camera.SOLUTION: The coloration reaction detection system comprises: an observation angle calculation part for calculating an observation angle of an imaging direction and a reference line of an observation object face with captured image data that are obtained by imaging a coloration reaction appearance region in an analysis kit; an image conversion part for converting the observation angle of the captured image data and generating captured image data for detection similar to an observation angle of reference captured image data indicating a reference pattern for discriminating an inspection result; a similarity calculation part for comparing a pattern of the coloration reaction appearance region with the reference pattern of the reference captured image data to calculate similarity between the pattern and the reference pattern; a coloration reaction detection part for extracting a reference pattern that is most similar to the coloration reaction pattern; and an output part which outputs an inspection result corresponding to the extracted reference pattern.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a color reaction detection system, a color reaction detection method, and a program for detecting a color reaction of a simple analysis kit using a simple test paper or a simple test drug used in a pregnancy test or an influenza test. .

  In recent years, simple analysis kits using simple test papers, simple test drugs, and the like have been frequently used for determining diseases, pregnancy, chemical substances, and the like. For example, as an example of a simple analysis kit, a general example of a simple analysis method using immunochromatography will be described below. When an analyte containing an antigen is dropped onto one end of a strip formed of a cellulose film, the analyte dissolves the reagent in the strip. The antigen moves through the strip by capillary action while forming an immune complex with the labeled antibody labeled with a metal colloid or the like in the reagent. The antigen that has formed the immune complex is trapped on a capture antibody that is placed in a specific area (eg, an area on the line) in the strip. When this immune complex is trapped, a color reaction occurs, and the line resulting from this color reaction is visually confirmed, and the test result for the subject is analyzed.

  In addition, immunochromatography includes an analysis method, such as a densitometric analysis method, in which the spectral absorbance of a portion where a color reaction has occurred after separation is measured, and the concentration of a target substance in a specimen is measured.

As described above, a dedicated device for performing analysis using immunochromatography and a detection method using this dedicated device have already been used.
However, the dedicated device used for this analysis is a device for professionals (medical workers), and is generally expensive and can be used only in a very limited place.
In addition, the detection method using the dedicated device described above is premised on that illumination is performed using a special optical device incorporated in the dedicated device or photographing is performed in a predetermined environment (for example, Patent Document 1). reference).

  The above-described analysis method is widely used as a simple test kit that can be used at home in pregnancy tests and influenza tests. In recent years, “Point of Care Testing (POCT)” has been advocated, and biochemical test items that could only be processed by specialized hospital equipment can be used at low cost, using simple equipment anytime and anywhere, even without expert knowledge. There is an urgent need to create a mechanism for easily testing, and there is a tendency to make testing more accessible to patients. In addition, a test result can be obtained immediately at a medical site, and a merit that a doctor can perform a quick treatment is born. By using this simple test kit, it is possible to easily perform a preliminary diagnosis in an ambulance or at home even outside the medical site, and it is possible to efficiently deal with the medical site after being transported to the hospital.

JP 2011-214863 A

  However, the analysis method described above cannot be easily applied to an image captured by a general-purpose camera that has inferior imaging performance (luminance gradation resolution, sensor sensitivity, etc.) in a home environment including a light source as disturbance. .

  In the simple test kit described above, the test items corresponding to POCT include, for example, arterial blood gas analysis, hemoglobin test, cholesterol level measurement, electrolyte measurement, blood glucose measurement, infectious disease test, heart disease marker, urine test, fecal occult blood There are tests, blood coagulation tests, and H. pylori breath tests. Inspection items are expected to expand and increase in the future. In addition, it is thought that multi-functionalization in which multiple types of items are inspected with one simple analysis kit will progress.

In this multifunctional simplified immunochromatography analysis kit, there may be a plurality of test lines (for example, n) including a control line (or reference line) indicating that the immune complex has passed through the test region. Conceivable.
In this case, since there are n test lines indicating the color reaction, the combination of the presence / absence of the color reaction in the n test lines is considered in consideration of the two determination results of the presence / absence of the color reaction in the test line. There are 2 ⁿ patterns. Therefore, when the number of test items increases, the combination pattern of the presence / absence of a color reaction on the test line increases exponentially, and the inspection result also corresponds to the number.

It is more difficult to create an instruction manual indicating the inspection result for each combination pattern as the number of the above-described combination patterns increases.
Even if this instruction manual is created, it is difficult to find out the result of the inspection performed by itself from a huge number of patterns and confirm the determination result. Furthermore, there is a possibility that the user's own pattern is determined as a different pattern in the instruction manual and the inspection result is confirmed to make a mistake.

  The present invention has been made in view of such a situation, and does not use a special and expensive inspection device, and uses a plurality of inspection items of the analysis kit based on images captured by a simple imaging device such as a general-purpose digital camera. The present invention provides a color reaction detection system, a color reaction detection method, and a program capable of determining the pattern and easily obtaining inspection results for a plurality of inspection items.

  In order to solve the above-described problem, the color reaction detection system of the present invention is a detection system that detects a line generated by a color reaction of a target substance to be detected in an analysis kit using immunochromatography. The captured image data obtained by capturing the observation target surface including the color reaction appearance region where the line of the analysis kit appears, and the imaging direction in which the captured image data is captured and the reference line of the observation target surface An observation angle calculation unit for obtaining an observation angle formed, and a detection captured image similar to the observation angle of the reference captured image data in which the reference angle of the line for converting the observation angle of the captured image data and determining the inspection result is shown A ratio between the image conversion unit that generates data, the color reaction pattern of the line of the color reaction appearance region, and the reference pattern of the reference captured image data. And a similarity calculation unit that calculates the similarity between the color reaction pattern and the reference pattern, and the most similar to the color reaction pattern among the plurality of reference patterns for which the similarity is calculated. A color reaction detection unit that extracts a reference pattern having a high value and an output unit (for example, the display unit 108 in the embodiment) that outputs a test result corresponding to the extracted reference pattern.

  In the color reaction detection system of the present invention, the color reaction detection system detects the region to be detected by detecting a region generated by the color reaction of the material to be detected in the sample in the analysis kit. The captured image data is image data obtained by imaging an observation target surface including the color reaction appearance region where the color reaction appears in the analysis kit, and the reference pattern represents the color reaction. It is a pattern for determining the inspection result of the region that appears, and the similarity calculation unit compares the color reaction pattern of the region of the color reaction appearance region with the reference pattern of the reference captured image data. Features.

  The color reaction detection system of the present invention determines whether or not the imaging angle of the captured image data is included in a preset imaging angle range, and the imaging angle is included in the imaging angle range. If not, the image pickup control unit is further provided with an image pickup control unit that performs notification for prompting to pick up a picked-up image again by changing the image pickup angle.

  In the color reaction detection system of the present invention, the imaging control unit refers to a luminance value in a predetermined reference region of the analysis kit in the reference image data imaged under the same light source as that previously captured the reference pattern The brightness value is compared with the brightness value at the same position as the predetermined position of the captured image data, and the brightness value of the captured image data is corrected so that the brightness value and the reference brightness value are the same. It is characterized by that.

  The color reaction detection system according to the present invention provides color viewing between the color correction patch of the analysis kit in the captured image data and the color viewing patch captured under the same light source as that previously captured the reference image data. And an exposure control unit that adjusts the color of the captured image data so that the color correction patch of the captured image data is the same as the patch of color reference of the reference image. To do.

  The color reaction detection system of the present invention is characterized in that the analysis kit is configured to detect a specimen by a color reaction of a detection target substance using an immunochromatography method.

  In the color reaction detection system of the present invention, the analysis kit is a simple medium of at least pH test paper, urine test paper, and oxidation degree measurement paper as a medium for detecting a sample by a color reaction of a substance to be detected. An inspection paper is used.

  The color reaction detection method of the present invention is a detection method for detecting a line generated by a color reaction of a target substance to be detected in an analysis kit using immunochromatography, and the observation angle calculation unit is the analysis kit. The observation angle formed by the imaging direction in which the captured image data is captured by the captured image data obtained by capturing the observation target surface including the color reaction appearance region where the line appears, and the reference line of the observation target surface The observation angle calculation process for obtaining the detection angle, and the image conversion unit converts the observation angle of the captured image data and detects the same as the observation angle of the reference captured image data in which the reference pattern of the line for determining the inspection result is shown. An image conversion process for generating captured image data, a similarity calculation unit, a color reaction pattern of a line in the color reaction appearance region, and a reference to the reference captured image data The similarity calculation process of calculating the similarity between the color reaction pattern and the reference pattern by comparing the turn, and the color reaction detection unit, from among the plurality of reference patterns calculated the similarity, A color reaction detection process for extracting a reference pattern having the highest similarity to the color reaction pattern, and an output unit including an output process for outputting a test result corresponding to the extracted reference pattern. Features.

  The program of the present invention is a program for causing a computer to execute the function of a color reaction detection system that detects a line generated by a color reaction of a target substance to be detected in an analysis kit using immunochromatography. The imaging direction in which the captured image data is captured by the captured image data obtained by capturing the observation target surface including the color reaction appearance region where the line of the analysis kit appears, and the reference line of the observation target surface An observation angle calculation means for obtaining an observation angle formed by the above-mentioned: an imaging for detection similar to the observation angle of the reference captured image data in which the reference angle of the line for determining the inspection result is shown by converting the observation angle of the captured image data Image conversion means for generating image data, a color reaction pattern of the color reaction appearance region line, and the reference imaging Similarity calculating means for comparing a reference pattern of image data and calculating a similarity between the color reaction pattern and the reference pattern, and the color reaction from among the plurality of reference patterns for which the similarity is calculated This is a program for functioning as a color reaction detection means for extracting a reference pattern having the highest similarity to a pattern and an output means for outputting a test result corresponding to the extracted reference pattern.

  As described above, according to the present invention, determination of patterns of a plurality of inspection items of an analysis kit is performed based on an image captured by a simple imaging device such as a general-purpose digital camera without using a special and expensive inspection device. It is possible to provide a color reaction detection system, a color reaction detection method, and a program capable of easily obtaining inspection results for each of a plurality of inspection items.

It is a block diagram which shows the structural example of the color reaction detection system by the 1st Embodiment of this invention. It is a figure which shows the external appearance of the analysis kit using the immunochromatography method which is an example of the analysis kit of the analysis object of the color reaction detection system in this embodiment. It is a figure explaining the observation angle of the imaging part 101 at the time of imaging a color reaction appearance area | region. It is a figure which shows the structural example of the table which shows a response | compatibility with the reference pattern previously stored in the image data memory | storage part 109, and a test result. It is a flowchart which shows the operation example of the color reaction detection process with respect to the color reaction detection target using the simple test paper and simple test chemical in the detection system of 1st Embodiment. It is a block diagram which shows the structural example of the color reaction detection system by 2nd Embodiment.

<First Embodiment>
A first embodiment of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing a configuration example of a color reaction detection system according to the first embodiment of the present invention. In FIG. 1, the color reaction detection system 1 includes an imaging unit 101, an exposure control unit 102, an observation angle calculation unit 103, an imaging control unit 104, an image conversion unit 105, a similarity calculation unit 106, a color reaction detection unit 107, Each of the display unit 108 and the image data storage unit 109 is provided.

  The imaging unit 101 is, for example, a digital camera using an image sensor such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor), or an imaging device provided in a portable terminal. Output captured image data. In the first embodiment, the captured image data is a monotone image such as black and white.

Here, the imaging unit 101 does not have a high brightness sensitivity resolution of a general-purpose digital camera. In the general-purpose digital camera, each pixel of the captured image data is recorded with, for example, 256 gradations of brightness sensitivity resolution from 0 to 255 gradations. For this reason, when the captured image data having the gradation necessary for performing the densitometry analysis method is acquired in the analysis kit, it is impossible with only one captured image data.
For this reason, as a technique for artificially increasing the brightness sensitivity resolution with a general-purpose digital camera, the shutter speed is changed under the control of the exposure control unit 102, and a plurality of captured image data with different exposure conditions is captured. There is a method of combining and using image data. As an example of this method, one of the well-known methods is HDR synthesis method by P. Debevec (Paul E. Debevec and Jitendra Malik. Recovering High Dynamic Range Radiance Maps from Photographs. In SIGGRAPH 97, August 1997 ) Can be applied.

  When the imaging unit 101 captures captured image data that is a captured image of the analysis kit, the imaging control unit 104 captures imaging conditions of the imaging unit 101 such as a depth of focus and sensitivity (ISO (International Organization for Standardization) sensitivity) of the imaging device. Is controlled to a preset numerical value.

  The exposure control unit 102 controls imaging conditions of the imaging unit 101 such as shutter speed, aperture value, presence / absence of illumination light, and intensity of illumination light as exposure imaging conditions. Further, the exposure control unit 102 corresponds to the brightness around the color reaction appearance region (analysis kit) imaged by the color reaction detection system 1, and emits illumination light such as a flash light source as necessary at the time of imaging. A lighting instruction is output to an illumination unit (not shown).

  FIG. 2 is a diagram showing the appearance of an analysis kit using an immunochromatography method, which is an example of an analysis kit to be analyzed by the color reaction detection system in the present embodiment. The analysis kit to be detected by this embodiment is, for example, an analysis kit container 900 having an appearance shown in FIG. The analysis kit container 900 includes an observation target surface 901 and a dropping unit 903. The observation target surface 901 includes a color reaction appearance region 902 and a labeled antibody reaction unit 904. The labeled antibody reaction unit 904 is a region where a labeled antibody that labels the detection target substance (antigen) contained in the specimen dropped into the dropping unit 903 is present. In the labeled antibody reaction unit 904, the labeled antibody reacts and binds to the detection target corresponding to each of the labeled antibodies to form an immune complex. In the present embodiment, the color reaction region in the color reaction appearance region 902 is defined as a region on the line. The immune complex moves in the color reaction appearance region 902 on the observation target surface 901, and reaches the control line 909 via the test lines 905, 906, 907, and 908. The control line 909 is provided to notify that the sample has passed all the test lines. The control line notifies the user of the end of the examination by stopping the movement of the specimen and indicating a color reaction.

  Each of the test lines 905, 906, 907, and 908 is an area where capture antibodies for different substances to be detected are fixed. The target substance to be detected corresponding to the capture antibody binds to the capture antibody and binds to the capture antibody in the test line (by the antigen-antibody reaction). As a result, if there is a target substance to be detected in the specimen, a test line with a capture antibody corresponding to the target object is detected by a color reaction by a metal colloid (coloring body) attached as a label to the labeled antibody. , It is recognized by coloring as a line of a predetermined color.

For example, since each of the test lines 906 and 907 includes a detection target corresponding to each capture antibody in the specimen, a color reaction line pattern is generated by the color reaction. Here, the color reaction line indicates a color line formed by binding a detection substance (that is, immune complex) corresponding to the capture antibody to the capture antibody of the test line. On the other hand, since each of the test lines 905 and 908 does not include the detection target corresponding to each capture antibody in the specimen, a line pattern due to the color reaction does not occur.
In the color reaction detection system 1 according to the present embodiment, the color reaction pattern formed by the color reaction line by the above-described color reaction and the test line that does not cause the color reaction is referred to in advance as a result of inspection. Each pattern group is compared, and the test result corresponding to the matching reference pattern is used as the test result of the specimen.

  The observation angle calculation unit 103 includes an observation position (coordinate value) that is a position at which imaging is performed in a three-dimensional space in which captured image data of the color reaction appearance region (color reaction appearance region 902 in FIG. 2) is captured. Each of the imaging directions of the imaging unit 101 is obtained from a coordinate conversion formula (described later). Here, the observation position indicates a coordinate value in a three-dimensional coordinate system indicating a three-dimensional space. That is, the observation angle calculation unit 103 determines the observation angle (or imaging angle) of the color reaction appearance region 902 (that is, the observation target surface 901 of the analysis kit 900) in each captured image data from the obtained observation position and imaging direction. Ask.

In the present embodiment, a color reaction appearance region 902 including a color reaction appearance region 902 is imaged by the imaging unit 101 with captured image data at a predetermined focal length set in advance (hereinafter, a color reaction appearance region). 902 is taken as an image), and taken image data.
From the captured image data, the observation angle calculation unit 103 calculates the observation angle of the captured image data obtained by capturing the color reaction appearance region 902 in the three-dimensional space by using a preset coordinate conversion formula as described above. Estimated.

  The coordinate conversion formula for obtaining the observation angle described above is used for preprocessing (preparation for performing color reaction detection) for performing color reaction detection on the color reaction appearance region 902 provided in the analysis kit. In this preprocessing, when a three-dimensional space is reproduced from a plurality of pieces of captured image data (captured image data obtained by imaging a calibration board described later), the pixel positions and three-dimensional coordinates of the plurality of captured image data are reproduced. It is an expression generated when associating coordinate positions in space. The coordinate conversion formula generated in advance is written and stored in advance in the image data storage unit 109 for each color reaction detection target. Here, the color reaction detection target indicates the type of analysis kit. When the shapes of the analysis kit containers 900 of the analysis kit are different, it is necessary to prepare coordinate conversion formulas generated according to the respective container shapes.

  FIG. 3 is a diagram illustrating an observation angle of the imaging unit 101 when imaging a color reaction appearance region. In FIG. 3, the sample is dropped from the dropping unit 903, and the sample moves in the color reaction appearance region 902 from the dropping unit 903 toward the control line 907 according to the principle of chromatography. If there is an object to be detected in this sample, the labeled antibody is bound in a labeled antibody reaction part (not shown) to generate an immune complex. Then, by binding to a capture antibody at a specific location (test line) in the color reaction appearance region 902, a color reaction line due to the occurrence of a color reaction caused by the labeling substance of the label antibody is observed.

The normal line 400 is a reference line for indicating the observation angle in the imaging direction of the captured image data, and is a normal line indicating the surface direction of the surface of the color reaction appearance region 902. The observation angle α is an angle formed by the imaging direction 101A of the imaging unit 101 and the normal line 400.
Here, for example, the observation angle calculation unit 103 performs analysis so that the direction parallel to the normal line 400 is the z-axis and each side of the color reaction appearance region 902 is parallel to each of the x-axis and the y-axis. The kit is placed in a three-dimensional coordinate system. For example, in the three-dimensional coordinate system, the analysis kit 500 is made up of the x axis and the y axis so that any one of the vertices formed by each side of the analysis kit container 900 coincides with the origin O of the three-dimensional coordinate system. Place in the dimension plane. For this reason, the thickness direction of the analysis kit container 900 is parallel to the z-axis. The three-dimensional shape of the analysis kit container 900 is stored in advance in the image data storage unit 109 as known information together with the coordinate conversion formula already described.

  Returning to FIG. 1, when the observation angle calculation unit 103 obtains the observation angle of the captured image data, the coordinates of the three-dimensional shape of the analysis kit container 900 in the three-dimensional coordinate system and the captured image data (two-dimensional coordinate system). Each pixel (coordinate) is associated with the coordinate conversion formula. And the observation angle calculation part 103 calculates | requires the imaging position of the captured image data from this imaging position, and the imaging position of the captured image data in the three-dimensional coordinate system of three-dimensional space by the said correlation. At this time, as described above, the observation angle calculation unit 103 has one vertex of the three-dimensional shape of the color reaction appearance region 902 in the three-dimensional coordinate system as the origin, and the normal 400 is parallel to the z axis. A three-dimensional space is defined on the analysis kit container 900 so that each side is parallel to the x-axis or the y-axis.

  Then, the observation angle calculation unit 103 obtains the imaging position and imaging direction of the captured image data of the imaging unit 101 in the three-dimensional coordinate system with reference to the three-dimensional shape of the analysis kit container 900. Thereby, the observation angle calculation unit 103 obtains an observation angle α formed by the normal 400 and the imaging direction 101A of the imaging unit 101.

  In the present embodiment, it is necessary on the premise that camera calibration (camera calibration) is performed on the imaging unit 101 in advance. In this camera calibration, a calibration board having a known three-dimensional shape is imaged once or a plurality of times in an imaging region, and the one or a plurality of captured image data is used to capture a three-dimensional space in a three-dimensional space. A coordinate point in the coordinate system is associated with a plurality of coordinate points (two-dimensional pixels) in the two-dimensional coordinate system of the captured image data. As a result, the coordinate conversion formula indicating the relative positional relationship between the imaging unit 101 and the calibration board (hereinafter, external parameters), the optical center of the imaging unit 101 and the light incident direction vector in each pixel (two-dimensional pixel), lens A distortion or the like (hereinafter, an internal parameter of the imaging unit 101) is estimated.

  Here, in this embodiment, in order that the observation angle calculation part 103 mentioned later estimates the observation angle of captured image data, the two-dimensional image which imaged the calibration board from several different viewpoint directions imaged with the imaging part 101 previously. That is, the global coordinate system (three-dimensional coordinate system) is reconstructed from the multi-viewpoint captured image data. Then, a coordinate conversion formula (internal parameter of the image capturing unit 10) indicating the correspondence between the coordinate point in the reconstructed 3D coordinate system in the same pixel and the coordinate point in the 2D coordinate system of the captured image data captured by the image capturing unit 101. The conversion formula obtained by (1) is obtained at the time of camera calibration.

  As described above, in this embodiment, the observation angle is estimated by performing camera calibration (camera calibration) on the imaging unit 101 in advance, so that the imaging unit is executed when the color reaction detection process is performed in the detection system. It is assumed that the internal parameters 101 are known and the three-dimensional shapes of the analysis kit container 900 and the color reaction appearance region 902 are known. As a result, the imaged image data of the color reaction appearance region 902 is imaged, and information on a plurality of corresponding points between the coordinate point in the three-dimensional coordinate system and the pixel in the two-dimensional coordinate system of the imaged image data is obtained by the coordinate conversion formula. The relative positional relationship between the imaging unit 101 and the color reaction appearance region 902 can be estimated from the corresponding point coordinates. That is, the observation position and observation angle (imaging direction) of the imaging unit 101 in the three-dimensional coordinate system when the color reaction appearance region 902 is imaged can be estimated.

  In this embodiment, for example, camera calibration is one of well-known techniques, such as Z.Zhang, “A flexible new technique for camera calibration”, IEEE Transactions on Pattern Analysis. and Machine Intelligence, Vol. 22, No. 11, pages 1330-1334, 2000) can be applied to estimate the observation angle when the captured image data is captured. However, when the observation angle is estimated by applying the analysis method according to Z.Zhang, the captured image data input to the detection system is the same focus (preferably the same focus) as the focus fixed during camera calibration. The image data must be captured at

  The image conversion unit 105 compares the color reaction pattern in the captured image data of the color reaction appearance area 902 with all the color reaction patterns (reference pattern groups) appearing in the color reaction appearance area 902 registered in advance. Therefore, the captured image data is normalized. Since all three-dimensional coordinates are known on the color reaction appearance region 902 in the captured image data, all the color reaction appearance regions 902 using the external parameters and internal parameters when the reference pattern group is imaged are used. If the three-dimensional coordinates are reprojected, the color reaction patterns in the color reaction appearance region 902 can be compared in the same coordinate system as the reference pattern group. That is, the image conversion unit 105 displays the color reaction appearance region 902 of the captured image data based on the obtained imaging angle α, and the reference color in the reference captured image data in which the reference pattern of the line for determining the inspection result is shown. It converts into the picked-up image data for a detection which is parallel and overlaps with the surface of a reaction appearance area | region. The detected captured image data is converted into an image captured from the same imaging position and imaging angle as the reference captured image data in the three-dimensional coordinate system.

  The similarity calculation unit 106 sequentially refers to the captured image data of the color reaction appearance region 902 and reference pattern group data registered in advance in the image data storage unit 109 to calculate the similarity. Since these images are normalized by the image conversion unit 105, the similarity calculation unit 106 performs template matching for all the pixels on the color reaction appearance region 902, and the score obtained thereby is used as the similarity. To do. Here, when the template matching is performed, the similarity calculation unit 106 calculates the luminance value for each pixel corresponding to each of the image data of the color reaction appearance region 902 of the captured image data and the image data of the reference pattern group, for example. Find the mean square error. Then, the similarity calculation unit 106 adds the above mean square error in all pixels (pixels) or some corresponding pixels, and outputs the addition result as a numerical value indicating the similarity. Therefore, the lower the similarity value, the more similar the captured image data and the reference pattern group data.

FIG. 4 is a diagram illustrating a configuration example of a table indicating the correspondence between the reference pattern stored in advance in the image data storage unit 109 and the inspection result.
FIG. 4A is a configuration example of a reference pattern table showing the correspondence between the reference pattern identification information and the reference pattern image index. The reference pattern identification information is identification information for identifying each reference pattern. The reference pattern image index indicates an address where the image of the reference pattern in the image data storage unit 109 is stored.
FIG. 4B is a configuration example of an inspection result table indicating correspondence between the reference pattern identification information and the inspection result indicated by the reference pattern. The inspection result indicates the result of the inspection corresponding to the detection target object bound to the capture antibody of the test line of the color reaction line.
FIG. 4C is a configuration example of a similarity table indicating the correspondence between the reference pattern identification information and the inspection result indicated by the reference pattern. The degree of similarity indicates the degree of similarity between the color reaction pattern in the detected captured image data and each reference pattern.

  Returning to FIG. 1, the similarity calculation unit 106 refers to the reference pattern table in the image data storage unit 109 and sequentially reads the reference pattern image index. Then, the similarity calculation unit 106 reads an image of the reference pattern from the image data storage unit 109 using the reference pattern image index. Then, the similarity calculation unit 106 obtains the similarity between the reference pattern sequentially read from the image data storage unit 109 and the color reaction pattern in the detected captured image data, and the obtained similarity is obtained in the image data storage unit 109. In the similarity table, the information is written and stored in the similarity column of the reference pattern identification information of the corresponding reference pattern.

The color reaction detection unit 107 refers to the similarity table stored in the image data storage unit 109, and from this similarity table, the color reaction pattern of the detected captured image data corresponding to all the reference pattern groups and Read the similarity of.
Then, the color reaction detection unit 107 compares each of the read similarities of all the reference pattern groups with a preset similarity threshold, and the similarity is the lowest among those below the similarity threshold (the most similar). Select the reference pattern.

The similarity threshold is a value that guarantees that the captured image of the color reaction appearance region matches one of the images of the reference pattern group, and is between each image data of the reference pattern group and the corresponding captured image data. It is set by using the similarity calculated in. It is considered that the difference in similarity between the image data of the reference pattern group and the color pattern reaction pattern of the detected captured image data that is the most similar to the other image data is significant. Therefore, there is usually only one reference pattern whose similarity is below the similarity threshold in the reference pattern group.
In addition, when there are a plurality of samples that are below the similarity threshold, the color reaction detection unit 107 selects the reference pattern with the lowest similarity among them. Then, the color reaction detection unit 107 refers to the inspection result table stored in the image data storage unit 109 and extracts the inspection result corresponding to the reference pattern identification information of the selected reference pattern.

  The display unit 108 is a display device such as a liquid crystal display for displaying, and displays the examination result extracted by the color reaction detection unit 107 on the display screen. Alternatively, the inspection result extracted by the color reaction detection unit 107 may be output from a printing apparatus such as a printer (not shown).

  Further, the imaging control unit 104 determines whether or not the observation angle at the time of imaging the color reaction appearance region 902 is within a preset angle range during imaging. Here, the angle range indicates a range of observation angles at which captured image data that can hold an amount of information capable of detecting a color reaction in the color reaction pattern when the image conversion unit 105 performs image conversion can be captured. . For example, when photographing is performed at an observation angle α (that is, α = 90 ° or 270 °) such that the observation direction is substantially orthogonal to the normal line 400 of the color reaction appearance region 902, the color reaction in the captured image data. The number of pixels in the image of the appearance area 902 becomes very small, the color reaction line becomes unclear, and the color reaction detection cannot be performed accurately.

Therefore, the imaging control unit 104 causes the observation angle calculation unit 103 to estimate the observation angle that is the imaging direction of the captured image data of the imaging unit 101.
When the observation angle estimated by the observation angle calculation unit 103 is within the angle range, the imaging control unit 104 satisfies the angle condition in the imaging process, while the estimated observation angle is not within the angle range. A display indicating that the angle condition in the imaging process is not satisfied is displayed on the display screen of the display unit 108 to prompt the user to adjust the observation angle within the angle range. At this time, the imaging control unit 104 displays on the display screen of the display unit 108 a notification notifying that it is moved slightly up and down and left and right. Alternatively, when imaging the imaging unit 101, the imaging control unit 104 sets a frame that matches the shape of the image of the analysis kit container 900 so that the display unit 108 has an observation angle that falls within an angle range on the display screen. indicate.

  In addition, the imaging control unit 104 sets imaging conditions for imaging captured image data having comparable quality with respect to the image data of the reference pattern group stored in the image data storage unit 109 when imaging. Determine if you are satisfied. Further, it may be configured such that the presence or absence of illumination (for example, a flash light source) or the intensity of illumination is added as necessary to the exposure conditions in the imaging conditions.

  Further, the imaging control unit 104 generates a luminance histogram when setting an exposure condition in the imaging unit 101 as an imaging condition. The imaging control unit 104 indicates the distribution of the gradation of each pixel, and the generated luminance in determining whether the distribution of the gradation in the captured image data is not biased toward the high gradation level or the low gradation level. A histogram is used. For example, when the distribution of gradation in the luminance histogram is biased toward the low gradation, that is, the gradation is expressed in 256 levels from “0” to “255”, and the gradation in the captured image data is “0”. When there are many neighboring pixels, blackout occurs in the captured image data, and comparison with the correct image data cannot be performed. On the other hand, when the distribution of the gradation in the luminance histogram is biased toward the high gradation, that is, when there are many pixels in the vicinity of the gradation “255” in the captured image data, whiteout occurs in the captured image data and the correct image data Cannot be compared.

For this reason, it is necessary to set the exposure conditions so that the distribution of the luminance histogram exists in the vicinity of the center of the range of gradation from “0” to “255”.
The imaging control unit 104 determines whether it is necessary to adjust the illumination or change the shutter speed based on the distribution of the gradation levels of the luminance histogram. When it is estimated that blackout will occur and the adjustment of illumination or shifting of the shutter speed is required to shift the distribution of the luminance histogram to the higher gradation level, the imaging control unit 104 instructs the exposure control unit 102 during imaging. The color reaction appearance region 200 is illuminated with a predetermined intensity (for example, flash light is irradiated in the imaging direction), or the brightness value is adjusted by increasing the shutter speed. In addition, when the color reaction detection system 1 does not have the exposure control unit 102 and illumination, the imaging control unit 104 irradiates the color reaction appearance region 200 with illumination having a light intensity necessary for the user. A display prompting the user is displayed on the display screen of the display unit 108.

Next, FIG. 5 is a flowchart illustrating an operation example of the color reaction detection process for the color reaction detection target using the simple test paper and the simple test medicine in the detection system of the first embodiment.
Step S101:
The imaging control unit 104 detects a current imaging condition of a color reaction detection target in the imaging unit 101, for example, an observation angle and an exposure condition.

Step S102:
The imaging control unit 104 determines whether or not all of the imaging conditions such as the exposure condition are imaging conditions that can capture the captured image data with image quality that can be compared with the image data of the reference pattern in the reference pattern group. Do.
At this time, the imaging control unit 104 advances the process to step S103 if the imaging condition allows imaging of captured image data with image quality that can be compared with the image data of the reference pattern in the reference pattern group. On the other hand, the imaging control unit 104 advances the process to step S <b> 104 when the imaging condition is not sufficient to capture the captured image data with image quality that can be compared with the image data of the reference pattern in the reference pattern group.

Step S103:
The imaging control unit 104 extracts the imaging position of the color reaction appearance region 902 in the captured image data. That is, the imaging control unit 104 obtains the three-dimensional shape of the analysis kit container 900 (color reaction detection target) within the imaging range of the imaging unit 101. Then, the imaging control unit 104 compares the obtained three-dimensional shape of the analysis kit container 900 with the three-dimensional shape of the reference analysis kit container 900 stored in advance, and within the imaging range of the imaging unit 101. A color reaction appearance region 902 region is extracted.

Step S104:
The imaging control unit 104 displays a condition that is not satisfied in the imaging condition on the display screen of the display unit 108, and suggests adjustment of the condition that is not satisfied in the imaging condition to the user.

Step S105:
The imaging control unit 104 compares the color reaction appearance region 902 in the imaging range of the imaging unit 101 with the color reaction appearance region 902 in the three-dimensional shape of the analysis kit container 900 stored in advance. Then, the imaging control unit 104 determines whether or not the entire color reaction appearance region 902 is included in the captured image data, that is, whether or not the entire color reaction appearance region 902 is included in the observation region. .
At this time, if the entire color reaction appearance region 902 is included in the captured image data, the imaging control unit 104 advances the process to step S106, while the entire color reaction appearance region 902 is included in the captured image data. If not, the process proceeds to step S107.

Step S106:
The imaging control unit 104 causes the observation angle calculation unit 103 to perform processing for estimating the imaging direction of the color reaction appearance region 902, that is, the observation angle.
As a result, the observation angle calculation unit 103 performs the three-dimensional shape of the analysis kit container 900 obtained from the captured image data in the imaging range of the imaging unit 101, and the three-dimensional analysis kit container 900 in a pre-stored three-dimensional coordinate system. The color reaction appearance region 902 observation angle is estimated by comparing the shape. Here, the observation angle calculation unit 103 obtains an imaging direction in which the imaging unit 101 images the analysis kit container 900 based on the comparison. Then, the observation angle calculation unit 103 has a surface to which the color reaction appearance region 902 of the analysis kit container 900 in the three-dimensional coordinate system is attached (a color reaction appearance region 902 on the upper surface or the lower surface of the analysis kit container 900 is attached. The angle formed by the normal of any one of the planes) and the imaging direction of the imaging unit 101 is obtained as an observation angle, and is output to the imaging control unit 104.

Step S107:
The imaging control unit 104 displays on the display screen of the display unit 108 that the imaging position captured by the imaging unit 101 is adjusted so that the entire color reaction appearance region 902 is included in the imaging range of the imaging unit 101. Then, the change of the imaging position is suggested to the user.

Step S108:
The imaging control unit 104 determines whether or not the entire color reaction appearance region 902 is in the imaging direction in which the captured image data is captured, that is, the observation angle is within a preset angle range.
At this time, if the observation angle of the imaging unit 101 is within the angle range, the imaging control unit 104 advances the process to step S109. On the other hand, if the observation angle of the imaging unit 101 is not within the angular range, the process proceeds to step S110. Proceed to

Step S109:
The imaging control unit 104 displays an image indicating that the color reaction appearance region 902 can be imaged on the display screen of the display unit 108 and prompts the user to image the color reaction appearance region 902.
Then, the user confirms the display screen and inputs an imaging instruction from the input means of the color reaction detection system 1.
Thereby, the imaging control unit 104 causes the imaging unit 101 to perform imaging processing to obtain captured image data.

Step S110:
The imaging control unit 104 displays on the display screen of the display unit 108 that the imaging direction of the imaging unit 101 is adjusted so that the observation angle of the imaging unit 101 falls within a preset angle range. It suggests the user to change the imaging direction.

Step S111:
Based on the imaging angle supplied from the observation angle calculation unit 103, the image conversion unit 105 displays the color reaction appearance region 902 of the captured image data in the reference captured image data in which the reference pattern of the line for determining the inspection result is indicated. It converts into the picked-up image data for a detection which is parallel and overlaps with the surface of a reference color reaction appearance area. That is, the image conversion unit 105 converts the imaging position and the observation angle of the captured image data, and detects the same imaging position and observation angle as the reference captured image data in which the reference pattern of the line for determining the inspection result is shown. The captured image data is generated.
As a result, the image conversion unit 105 can compare the color reaction appearance region 902 of the captured image data with the reference color reaction appearance region of the reference captured image data as the same imaging position and observation angle. Further, the image conversion unit 105 normalizes the captured image data so as to perform template matching with each image of the reference pattern in the reference pattern group of the image data storage unit 109, and generates captured image data for detection.

Step S112:
The similarity calculation unit 106 sequentially reads the reference pattern image data from the reference pattern group of the reference pattern table of the image data storage unit 109, and presents the read reference pattern and the detected captured image data in the color reaction appearance region 902. Template matching is performed with the color reaction pattern.
Here, the similarity calculation unit 106 performs template matching between all the reference patterns in the reference pattern group of the reference pattern table and the color reaction patterns in the color reaction appearance region 902 of the detection captured image data.
Then, the similarity calculation unit 106 uses the image data storage unit 109 to calculate the similarity between the reference pattern in the reference pattern group and the color reaction pattern in the color reaction appearance region 902 of the detected captured image data obtained by template matching. In the similarity table, the similarity between each reference pattern and the color reaction pattern is written and stored in association with the reference pattern identification information of each reference pattern.

Step S113:
The color reaction detection unit 107 refers to the similarity table of the image data storage unit 109, and from among the similarities corresponding to all the reference patterns in the similarity table, a reference pattern that exceeds a preset similarity threshold value. The presence / absence judgment is performed.
At this time, if there is a reference pattern that exceeds the similarity threshold, the color reaction detection unit 107 proceeds to step S114. On the other hand, if there is no reference pattern that exceeds the similarity threshold, the process proceeds to step S115.

Step S114:
When there are a plurality of reference patterns exceeding the similarity threshold in the similarity table, the color reaction detection unit 107 selects the reference pattern with the highest similarity. Then, the color reaction detection unit 107 extracts an inspection result corresponding to the reference pattern identification information of the selected reference pattern from the inspection result table of the image data storage unit 109. The color reaction detection unit 107 displays the inspection result extracted from the inspection result table on the display screen of the display unit 108 and ends the process.

Step S115:
The color reaction detection unit 107 determines that the reference pattern corresponding to the captured image of the color reaction appearance region 200 is not in the reference pattern group, and displays on the display unit 108 that the color reaction could not be detected. And exit.

  With the configuration described above, according to the present embodiment, the captured image data obtained by imaging the color reaction appearance region 902 is compared with all the color reaction patterns that appear in the color reaction appearance region 902 in the reference pattern group, and the specimen Performs color reaction detection for inspection. Therefore, according to the present embodiment, an expensive and special detection device is not used as in the prior art, and it depends on the installation state of the color reaction appearance region 902 (such as an analysis kit having a plurality of types of test lines). Easily detect the color reaction pattern in an analysis kit using simple test paper or simple test chemicals by using the captured image of the color reaction appearance region 902 with a simple image capturing device such as a general-purpose digital camera. In addition, it is possible to carry out with high accuracy.

Here, the analysis kit 900 to be analyzed by the color reaction detection system will be described in detail.
As illustrated in FIG. 2, the analysis kit 900 may have a color reaction appearance region 902 on a flat surface, a tube shape such as a cylindrical shape filled with a test agent, or a color reaction appearance. The medium in the region 902 may be in the form of paper such as simple test paper such as pH (peech etch) test paper (including litmus test paper), urine test paper, and oxidation degree measurement paper (for cooking oil test).
In addition, in an analysis kit having a color reaction appearance area on a plane, the color reaction area showing the color reaction on the observation symmetry plane 901 in the color reaction appearance area 902 may be the line shape shown in FIG. Alternatively, a dot shape or a spot shape may be used.

In the color reaction appearance region on the plane, as illustrated in FIG. 2, as illustrated in FIG. 2, an immunochromatography in which a target substance to be detected is dropped on the dropping unit 903 and the color reaction appears in each color reaction appearance region of the target substance. Analyze kits that perform inspections using the graphic method, and observe the color reaction by immersing (for example, affixed) plates on which the test paper and reagents are provided on the observation symmetry plane 901 in the target substance to be detected. An analysis kit may be used.
Some analysis kits have a tube-like color reaction appearance region such as a dropper, and the target substance to be detected is sucked from the water absorption part of the analysis kit, and the non-detection target in the tube Some of them see a color reaction in response to immersion in a filling (test agent, etc.) in the tube. Any of these may be sufficient as the analysis kit in this invention, and you may be comprised by these combinations.

  On the other hand, in the tube-type analysis kit, the shape of the color reaction region may not be a plane as in the color reaction appearance region 902, but may be a cylinder (column). However, even in the case of an analysis kit having a shape on a tube-type cylinder, assuming that there is a pattern (region) whose three-dimensional coordinates are known on the surface of the cylindrical packing in the tube, The observation angle can be estimated as in the case of a plane. In other words, by taking the correspondence between the three-dimensional coordinates of the pattern (region) and the two-dimensional pixel coordinates on the captured image, the observation angle calculation unit 103 is used to display the color on the observation symmetry plane 901 on the plane already described. The observation angle can be estimated as in the case of detecting the color reaction region indicating the reaction.

In the analysis kit 900 described above, the target substance to be detected may be a specific protein that is a typical target substance to be detected using an immunochromatography method, or may be a microorganism such as a virus, a fungus, or a bacterium.
Further, the analysis kit 900 described above may be a nucleic acid that is detected in order for the substance to be detected to extract a specific DNA structure.
Further, in the above-described analysis kit 900, the target substance to be detected is a chemical substance (for example, a saccharide contained in food, a substance showing an oil oxidation degree for detecting a deterioration state of oil, a mixed substance mixed in a water quality test, or the like. , Measurement of sugar content based on sucrose content in fruits).

In the analysis kit 900, the color reaction detection system may detect a color change (change to a different type of color) in the color reaction appearance region 902 on the observation symmetry plane 901, or a color density change (the same color). It is also possible to detect a change in the concentration of (a).
Further, the color reaction detection system may be one that captures fluorescence or phosphorescence in the color reaction appearance region 902 in the analysis kit 900 and detects a color change (change to a different type of color) in the captured image. A color density change (change in the density of the same color) may be detected.

  In addition, when performing an inspection with an analysis kit for a target substance to be detected, the color reaction detection system captures a captured image as a moving image instead of a still image, and captures a change in the color reaction region, and is captured in a certain continuous time. It is also possible to detect the change speed of the color change or density change in the color reaction appearance region 902 from the moving image. Here, the color change indicates a change from an initial color to a preset color. As a result, the change time can be detected with higher accuracy than the measurer measures the color change or density change time.

<Second Embodiment>
Hereinafter, a second embodiment of the present invention will be described with reference to the drawings.
The second embodiment is the same as the configuration of the color reaction detection system 1 according to the first embodiment shown in FIG. Hereinafter, operations different from those of the first embodiment will be described. In the second embodiment, the captured image data is a monotone image such as black and white, for example, as in the first embodiment.
In the first embodiment, detection of a color reaction pattern based on the presence or absence of a color reaction line is targeted. However, the color reaction detection system 1 according to the second embodiment includes not only the presence / absence of the color reaction line but also the determination of the color reaction pattern including the detection of the density of the color reaction line.

In the case of light / dark determination, the light / light environment of the color reaction pattern causes the color reaction line of the color reaction pattern to be imaged at different gradations. For this reason, when comparing the density of the color reaction line of the color reaction pattern in the captured image data for detection with the density of the color reaction line of the reference pattern, an error occurs in the calculation of the similarity.
For this reason, in the same environment as the light source environment when the image data of the reference pattern in the reference pattern group is captured in advance, the surface of the analysis kit container 900, the surface of the labeled antibody reaction unit 904, and the surface of the base of the color reaction appearance region 902 The predetermined area is taken as a reference area, and the image data of the reference area is captured as correction image data, and the correction image data is written and stored in the image data storage unit 109.

Then, the imaging control unit 104 reads the correction image data from the image data storage unit 109 when imaging the captured image data. The imaging control unit 104 compares the gradation level (reference luminance value) of the read-out correction area image data with the gradation level (luminance value) of the image data of the reference area in the captured image data. Here, the gradation value of the reference area is, for example, an average value of the gradation degrees of each pixel in the reference area.
Here, the imaging control unit 104 divides the gradation level of the image data of the reference area in the captured image data by the gradation level of the correction area image data of the reference area, and sets the numerical value of the division result as a density correction coefficient. Then, the imaging control unit 104 multiplies the gradation level of each pixel of the captured image data by the above-described density correction coefficient to correct the gradation level of each pixel of the captured image data.
For example, the correction of the gradation of each pixel of the captured image data described above is performed by, for example, imaging processing of the color reaction appearance region in step S109 in the flowchart of FIG. 5, that is, image conversion processing in step S111 after the captured image data is captured. Done before.

  With the above-described configuration, the present embodiment obtains a density correction coefficient as a ratio of the gradation level of the image data in the reference area in the captured image data to the gradation level of the correction image data in the reference area, and uses this density correction coefficient as the captured image. The captured image data is corrected by multiplying the gradation level of each pixel of the data. For this reason, according to the present embodiment, it is possible to correct the gradation level of each pixel of the captured image data captured in a different light source environment in the same environment as the light source environment when the image data of the reference pattern is captured. it can. Thus, according to the present embodiment, when the template matching is used for comparison with the light and shade of the color reaction line of the reference pattern, the color reaction pattern of the captured image captured in a light source environment different from the light source environment of the reference pattern and The comparison process can be performed with high accuracy.

<Third Embodiment>
Hereinafter, a third embodiment of the present invention will be described with reference to the drawings.
The second embodiment is the same as the configuration of the color reaction detection system 1 according to the first embodiment shown in FIG. Hereinafter, operations different from those of the first embodiment will be described. In the third embodiment, the captured image data is a color image (including the shades of each color component RGB).
Using color images as captured image data and reference pattern image data, and comparing the color reaction pattern of the captured image data with the reference pattern to detect a color reaction, the color appearance of the captured image data is important. .

  The imaging control unit 104 analyzes the current captured image data, and performs a process of removing the influence of the light source environment on the captured image of the color reaction appearance region 902 of the analysis kit according to an instruction from the exposure control unit 102. When a general-purpose digital camera has a white balance function, the white balance function may be used to remove the influence of the light source environment.

  In addition, as a method for correcting the color (gradation degree of the color component RGB of each pixel), the following configuration can be used. A position on the observation target surface 901 of the analysis kit container 900 that does not affect the inspection of the surface of the color reaction appearance region 902 (for example, a distance farther than a position where a droplet part control line is generated when viewed from the labeled antibody reaction part 904) 3), three correction patches having different colors are arranged in advance. Here, the position that does not affect the inspection indicates, for example, an area other than the area where the test line from the labeled antibody reaction unit 904 to the control line in the color reaction appearance area 902 is arranged on the observation target surface 901. Yes.

A reference pattern is imaged in a standard light source (for example, D65 light source) environment, spectral reflectance is measured in advance from the image data of each correction patch imaged at the same time, and is written and stored in the image data storage unit 109. deep.
By using the spectral reflectance of each of the three correction patches, the color of the color reaction pattern in the captured image data is corrected to obtain the captured image data of the color captured in the standard light source environment. Can be obtained and compared as a reference pattern and a color reaction pattern captured in the same light source environment.

The color of the correction patch imaged with a standard light source is calculated as C _gt (R _gt , G _gt , B _gt ), and the color of the correction batch in the captured image data is determined as C _in (R _in , G _in , B _in ), the following formula (1) is established between the color C _gt and the color C _in when the simplest color correction conversion formula is applied as an example.

In the equation (1), as described above, C _gt is the color of the correction patch imaged with a standard light source, and is described as the color component RGB (R _gt , G _gt , B _gt ). , C _in is the color of the correction batch in the captured image data, and is described as the color component RGB (R _in , G _in , B _in ).
Thereby, (1) Formula is represented by the following (2) Formula.

By using three correction patches each having a different color, there are three or more combinations of the color of the correction patch imaged with a standard light source and the color of the correction batch in the captured image data. By obtaining, the matrix A in the equation (2) is obtained.
Here, the imaging control unit 104 reads out three types of correction patch colors C _gt (R _gt , G _gt , B _gt ) and the expression (2) from the image data storage unit 109. Then, the imaging control unit 104 extracts the color C _in (R _in , G _in , B _in ) of the three types of correction batches from the captured image data, and the color C _gt (R _gt , R _gt , G _gt , B _gt ) is used to calculate matrix A using equation (2) corresponding to each of the three types of correction patches. The imaging control unit 104 converts the color tone of each pixel of the captured image data captured under an arbitrary light source into a color imaged under a standard light source, using the calculated matrix A, using equation (2). To do.

As described above, the matrix A is obtained, and the color of the captured image data captured under an arbitrary light source is converted into the color captured under a standard light source by using Equation (2). It becomes possible. As a result, the reference pattern and the color reaction pattern can be compared in a unified state under a standard light source. Since the reference pattern and the color reaction pattern are compared with each other under a standard light source, the influence of illumination (under the light source environment) can be eliminated from the captured image data, and a highly accurate color reaction detection process can be performed. .
For example, the color correction processing for the captured image data described above is performed by imaging the color reaction appearance region 902 in step S109 in the flowchart of FIG. 5, that is, the captured image of the observation target surface 901 including the color reaction appearance region 902. After the data is imaged, this is performed before the image conversion process in step S111.

  In addition, as already described, the similarity calculation unit 106 sequentially refers to the reference pattern group data registered in advance in the image data storage unit 109 in step S112 to calculate the similarity. The captured image data for detection used here is normalized by the image conversion unit 105, and template matching is performed for all the pixels on the color reaction appearance region 902. When this template matching is performed, the similarity calculation unit 106, for example, for each color component RGB of each pixel corresponding to each of the image data of the color reaction appearance region 902 and the image data of the reference pattern group in the captured image data. The mean square error of the luminance value of is calculated. Then, the similarity calculation unit 106 adds the above mean square error in all pixels (pixels) or some corresponding pixels, and outputs the addition result as a numerical value indicating the similarity. Therefore, the lower the similarity value, the more similar the captured image data and the reference pattern group data.

  Further, the similarity calculation unit 106 converts the numerical values of the color components RGB of all or some of the corresponding pixels in the captured image data and the correct image data into an appropriate color space, and then calculates the Euclidean distance of the color space. A configuration may be adopted in which square values are added and the addition result is output as a numerical value indicating the degree of similarity. Also in this case, as in the case of using the mean square error, the lower the similarity value, the more similar the captured image data and the correct image data.

As described above, the similarity calculation unit 106 includes the captured image data of the color reaction appearance region 902 in the captured image data for detection, the image data of the reference pattern in the reference pattern group stored in the image data storage unit 109, and The degree of similarity is obtained sequentially.
Then, the similarity calculation unit 106 writes the similarity corresponding to each obtained reference pattern in the similarity table in the image data storage unit 109 in association with the reference pattern identification information of the reference pattern for which the similarity is obtained. Remember.

Further, when the intensity of light emitted from the light source when the captured image data is captured does not correspond to the image data of each reference pattern of the reference pattern group, the color reaction appearance region 902 of the captured image data and the reference pattern Simple pixel comparison with image data is not possible.
For this reason, the similarity calculation unit 106 evaluates the color component RGB between predetermined pixels, that is, R / G (R gradation and G gradation between predetermined pixels of the detection image data). Ratio) and the R / G between the pixels of the image data of the reference pattern group corresponding to the predetermined pixels of the detected captured image data may be calculated.

  The term “between predetermined pixels” refers to, for example, a pair of two pixels PA and PB in advance, and shows the correspondence between this pixel PA and pixel PB. Here, as a combination of pixels used between predetermined pixels, a combination of pixels in which R / G and B / G are increased is set in advance. Then, as a ratio obtained by dividing the R gradation of the pixel PA by the G gradation of the pixel PB, the pixel at the corresponding position in each of the color reaction pattern image data of the captured image data and the reference pattern image data R / G is obtained for each set. In addition to R / G, a combination of B / G (ratio between B gradation and G gradation) is also used in the same manner as R / G, and the similarity between the reference pattern and the color reaction pattern is determined. May be calculated.

The above-described color correction processing performed on the detected captured image data by the similarity calculation unit 106 may be combined with the above-described color correction processing performed on the captured image data by the imaging control unit 104. Only one of the color corrections may be performed.
As described above, according to the present embodiment, since color correction is performed on each pixel in a color image, it is possible to absorb a difference in light intensity under different light sources, and a reference pattern and a color reaction pattern It is possible to calculate a numerical value indicating a high degree of similarity in template matching.

<Fourth Embodiment>
Hereinafter, a second embodiment of the present invention will be described with reference to the drawings.
FIG. 6 is a block diagram illustrating a configuration example of a color reaction detection system according to the second embodiment. In FIG. 6, the color reaction detection system 1 </ b> A includes a color reaction detection device 10 and an imaging device 20. The color reaction detection device 10 includes an observation angle calculation unit 103, an imaging control unit 104, an image conversion unit 105, a similarity calculation unit 106, a color reaction detection unit 107, a display unit 108, and an image data storage unit 109. ing. The imaging device 20 includes an imaging unit 101 and an exposure control unit 102. In FIG. 6, the same components as those in the first embodiment are denoted by the same reference numerals.

In the present embodiment, the color reaction detection system is configured to be separated from the color reaction detection device 10 with the imaging and exposure functions in the first embodiment as the imaging device 20. As a result, a general-purpose digital camera or a portable terminal (including a mobile phone or a smart fan) can be easily used as the imaging device 20 for imaging image data for color reaction detection.
Moreover, although not shown in figure, the color reaction detection system 1A may communicate with a digital camera or a portable terminal using an information communication line such as the Internet. The color reaction detection device 10 is provided as a server on the cloud, and uses the captured image data transmitted from the digital camera or mobile terminal that is the imaging device 20 as in the first embodiment already described. The color reaction pattern detection process in the color reaction appearance region 902 may be performed.
The configuration of the fourth embodiment can be used as the configuration of the first to third embodiments already described.

Note that the program for realizing the functions of the color reaction detection system 1 of FIG. 1 and the color reaction detection system 1A of FIG. 6 according to the present invention is recorded on a computer-readable recording medium and recorded on the recording medium. The color reaction detection process may be performed by causing the computer system to read and execute the program. Here, the “computer system” includes an OS and hardware such as peripheral devices.
The “computer system” includes a WWW system having a homepage providing environment (or display environment). The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. Further, the “computer-readable recording medium” refers to a volatile memory (RAM) in a computer system that becomes a server or a client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. In addition, those holding programs for a certain period of time are also included.

  The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

DESCRIPTION OF SYMBOLS 1,1A ... Color reaction detection system 10 ... Color reaction detection apparatus 20 ... Imaging device 101 ... Imaging part 101A ... Imaging direction 102 ... Exposure control part 103 ... Observation angle calculation part 104 ... Imaging control part 105 ... Image conversion part 106 ... Similarity calculation unit 107 ... color reaction detection unit 108 ... display unit 109 ... image data storage unit 400 ... normal 900 ... analysis kit container 901 ... observation target surface 902 ... color reaction appearance region 903 ... dropping unit 904 ... label Antibody reaction part 905, 906, 907, 908 ... Test line 909 ... Control line

Claims (9)

A color reaction detection system that detects a target substance to be detected by a color reaction in an analysis kit,
The imaging direction in which the captured image data is captured by the captured image data of the observation target surface including the color reaction appearance region where the color reaction appears in the analysis kit, and the reference line of the observation target surface An observation angle calculation unit for obtaining an observation angle formed by:
An image conversion unit that converts the observation angle of the captured image data and generates detection captured image data similar to the observation angle of the reference captured image data in which a reference pattern of the color reaction for determining an examination result is shown; ,
A similarity calculation unit that compares a color reaction pattern of the color reaction appearance region with a reference pattern of the reference captured image data and calculates a similarity between the color reaction pattern and the reference pattern;
A color reaction detection unit that extracts a reference pattern having a high similarity to the color reaction pattern from the plurality of reference patterns for which the similarity is calculated;
A color reaction detection system, comprising: an output unit that outputs an inspection result corresponding to the extracted reference pattern. The color reaction detection system detects the detection target substance by detecting a region generated by the color reaction of the detection target substance of the sample in the analysis kit,
The captured image data is image data obtained by imaging the observation target surface including the color reaction appearance region where the color reaction in the analysis kit appears.
The reference pattern is a pattern for determining an inspection result of the region where the color reaction appears,
The color reaction detection system according to claim 1, wherein the similarity calculation unit compares the color reaction pattern of the color reaction appearance region and the reference pattern of the reference captured image data. It is determined whether or not the observation angle of the captured image data is included in a preset observation angle range. If the observation angle is not included in the observation angle range, the observation angle is changed again. The color reaction detection system according to claim 1, further comprising: an imaging control unit that performs notification that prompts the user to capture a captured image. The imaging control unit
A reference luminance value that is a luminance value in a predetermined reference region of the analysis kit in the reference image data imaged under the same light source as the image of the reference pattern in advance, and the same position as the predetermined position of the captured image data 4. The color reaction detection system according to claim 3, wherein the brightness value of the captured image data is corrected so that the brightness value of the position is compared and the brightness value and the reference brightness value are the same. . The color of the picked-up image data is compared with the color-seeking correction patch of the analysis kit in the picked-up image data and the color-seeking patch picked up under the same light source as the reference image data picked up in advance. 5. The exposure control unit according to claim 3, further comprising: an exposure control unit configured to adjust a color of the captured image data so that the viewing correction patch is similar to the color viewing patch of the reference image. Color reaction detection system. The analysis kit is
The color reaction detection system according to any one of claims 1 to 5, wherein the specimen is detected by a color reaction of a target substance to be detected using an immunochromatography method. The analysis kit is
2. A simple test paper such as a pH (pee etch) test paper, a urine test paper, or an oxidation degree measurement paper is used as a medium for detecting a specimen by a color reaction of a target substance to be detected. The color reaction detection system according to claim 5. A detection method for detecting a region generated by a color reaction of a target substance to be detected in a sample in an analysis kit,
The observation angle calculation unit uses the captured image data obtained by capturing the observation target surface including the color reaction appearance region where the region of the analysis kit appears, and the imaging direction in which the captured image data is captured, and the observation target An observation angle calculation process for obtaining an observation angle formed with the reference line of the surface,
Image conversion in which the image conversion unit converts the observation angle of the captured image data and generates captured image data for detection similar to the observation angle of the reference captured image data in which the reference pattern of the region for determining the inspection result is indicated Process,
The similarity calculation unit compares the color reaction pattern in the color reaction appearance region and the reference pattern of the reference captured image data, and calculates the similarity between the color reaction pattern and the reference pattern. Similarity calculation process,
A color reaction detection process in which a color reaction detection unit extracts a reference pattern having the highest similarity to the color reaction pattern from among the plurality of reference patterns for which the similarity is calculated,
The color reaction detection method, wherein the output unit includes an output process of outputting the inspection result corresponding to the extracted reference pattern. A program that causes a computer to execute a function of a color reaction detection system that detects a region generated by a color reaction of a target substance to be detected in an analysis kit,
The computer,
The captured image data obtained by capturing the observation target surface including the color reaction appearance region where the region of the analysis kit appears, and the imaging direction in which the captured image data is captured and the reference line of the observation target surface An observation angle calculation means for obtaining an observation angle formed;
Image conversion means for converting the observation angle of the captured image data and generating captured image data for detection similar to the observation angle of the reference captured image data in which a reference pattern of a region for determining an inspection result is indicated;
A similarity calculation means for comparing the color reaction pattern of the color reaction appearance region and the reference pattern of the reference captured image data, and calculating the similarity between the color reaction pattern and the reference pattern;
Color reaction detecting means for extracting a reference pattern having the highest similarity to the color reaction pattern from the plurality of reference patterns for which the similarity is calculated,
A program for functioning as an output means for outputting an inspection result corresponding to the extracted reference pattern.

Images