JP2023034611A - Image reading system - Google Patents

Abstract

To solve a problem in which: sometimes a document is stopped while being conveyed, and a read image is elongated in an image reading device that conveys the document and reads the document with a fixed reading sensor, and at this time, the elongation of the image is not constant, and it is necessary to accurately measure and correct the elongation of the document.MEANS FOR SOLVING THE PROBLEM: In an image reading system having an image reading device that reads an image of a document through a surface of a carrier sheet conveyed with a document inserted therein, a linear mark M, which is provided on the back surface of the carrier sheet and whose coordinate position in the width direction orthogonal to the conveying direction changes continuously at the same inclination with respect to the conveying direction in which the carrier sheet is conveyed, is read, and scaling processing on the image of the document is performed on the basis of a result of reading the mark M.SELECTED DRAWING: Figure 1

Description

The present invention relates to an image reading system having an image reading device.

Some image reading apparatuses have a configuration in which a reading sensor is fixed and a document to be read is read while being conveyed. In such a form, if the speed at which the document is conveyed and the reading timing of the reading sensor do not match, the reading result will be incorrect.

For example, when reading a document such as a passport that is much thicker than a normal document, the document may get caught in the transport path or the like during transport, and the transport may stop in some cases. Alternatively, if the speed at which the document is conveyed becomes slower than the preset speed, the read image will be stretched. FIG. 9 is an example of an image when a passport is read. In FIG. 9, C is an image when it is read normally, and D is an example of an image stretched because the document is caught or stopped during transportation. . If the image is stretched in this way, it will adversely affect the appearance and the recognition of the written content.

As a countermeasure against the read image being stretched, marks are placed on the carrier sheet at regular intervals, the marks are read, and from the results, variations in the speed at which the document is conveyed are detected, and the image is corrected to an appropriate magnification. There is (for example, see Patent Document 1).

JP-A-9-270907

However, when marks are provided at regular intervals as in Patent Document 1, the conveying speed is detected by the length of the marks and correction processing is performed. In this case, only the average speed between one mark portion can be measured, and correct correction cannot be performed unless the speed is constant in one mark portion. In this state, the magnification of the entire document area can be roughly corrected, but the stretching of the image cannot be correctly corrected where the transport speed actually changes.

In view of the above, the image reading system according to the present invention is
In an image reading system having an image reading device that reads an image of a document through a surface of a carrier sheet conveyed with the document sandwiched therebetween,
reading a linear mark provided on the back surface of the carrier sheet and having coordinate positions in the width direction perpendicular to the conveying direction continuously changing at the same inclination with respect to the conveying direction in which the carrier sheet is conveyed; It is characterized in that enlargement/reduction processing is performed on the image of the document based on the reading result.

According to the present invention, it is possible to correct the image of the document to an appropriate magnification.

FIG. 4 is a diagram showing a state in which a document is sandwiched between carrier sheets according to the present invention; FIG. 2 is a configuration diagram of an image reading unit according to the present invention; 1 is a configuration diagram of an entire image reading apparatus according to the present invention; FIG. 4 is a flowchart of document area detection processing in the image processing unit according to the present invention; 4 is a flowchart of diagonal line detection processing of a carrier sheet in an image processing section according to the present invention; 4 is a flowchart of image correction processing in the image processing unit according to the present invention; FIG. 4 is an explanatory diagram of image correction processing according to the present invention; FIG. 4 is an explanatory diagram of image correction processing according to the present invention; An example of an image that has been stretched due to stopping during document feeding.

An image reading apparatus according to a first embodiment of the present invention will be described below. FIG. 1 is a diagram showing an example of a state in which a document is sandwiched between carrier sheets used in an image reading apparatus according to this embodiment.

The carrier sheet is of a double-fold type, and A in FIG. 1 shows a state in which the front transparent surface is viewed from the front, and the document sandwiched between the carrier sheets can be seen through. Since B is an opaque surface on the back side, the sandwiched document is not transparent. In the following description, the A side is the front side and the B side is the back side.

A diagonal line is drawn on the back surface. At the time of image reading, the document is set on the conveying path with the front side facing downward while sandwiched between the carrier sheets.

Next, FIG. 2 is a schematic cross-sectional view of the image reading apparatus according to this embodiment. FIG. 3 is a control block diagram of the image reading apparatus according to this embodiment. In this embodiment, the image processing described below is performed by the image processing unit 103, which will be described later. However, the present invention is not limited to this. It may be configured as an image reading system in which image processing is executed by an information processing apparatus that receives an image output from the unit 104 . In that case, the image reading start instruction may be transmitted from the information processing apparatus.

As shown in FIG. 2, in the image reading apparatus 100 of the present embodiment, a paper feed tray 2 for stacking originals (sheets) is supported by an apparatus main body 1A via a hinge portion 2a so as to be openable and closable. That is, the paper feed tray 2 is configured to cover the document feed port 13 in the closed state (broken line in FIG. 2), and to stack the documents on the stacking surface 2b in the open state (solid line in FIG. 2). be done.

A plurality of documents stacked on the paper feed tray 2 are separated and fed to the transport path one by one from the lowest document by the feed roller 4 and the separation pad 5 . The document fed to the transport path is transported downstream by the transport roller pair 6 . Further, images on both sides of the document are read by image reading units (image reading means) 8 and 9 , and then the document is discharged out of the apparatus by a conveying roller pair 7 .

The image reading apparatus 100 configured as described above is connected to an information processing apparatus such as a computer, and the reading operation is controlled by an operation from the information processing apparatus. The connection between the image reading device 100 and the information processing device can be established by, for example, a USB cable, SCSI cable, LAN cable, wireless communication, or the like.

Further, in the information processing apparatus to which the image reading apparatus of the present embodiment is connected, the image reading apparatus 100 can be controlled by executing software (control program) that enables the image reading apparatus 100 to be controlled. Become.

In practice, the user can set the image reading conditions such as the resolution, the number of colors, the number of gradations, the reading surface, and the reading area from the control screen of the image reading apparatus 100 displayed by executing the software. It may be made configurable. In this case, the image reading apparatus 100 performs the reading operation according to the setting conditions specified by the user. Only some of these settings may be configured to be settable. Also, this control program may be a general-purpose scan program that the OS (Operating System) of the information processing apparatus has.

Further, the image reading apparatus 100 transmits image data read by the image reading units 8 and 9 . For example, as shown in FIG. 3, in the image reading apparatus of the present embodiment, the control unit 102 comprising a CPU controls the image data read by the image reading units 8 and 9 to be transmitted to the outside by the image output unit 104. It is designed to

Here, although not shown, the image reading units 8 and 9 are composed of a light irradiation unit such as an LED for irradiating a document being conveyed with light of a plurality of colors, and a light receiving unit for receiving reflected light from the document. In this embodiment, the image reading units 8 and 9 irradiate light of R (red), G (green), and B (blue) as light of a plurality of colors to irradiate the original, and light reflected from the original is detected. It is assumed that one line is read. The control unit 102 controls the timing of light irradiation by such a light irradiation unit. The image reading units 8 and 9 are arranged across the transport path so as to be able to read the front and back of the original, respectively, and are attached so that the reading positions thereof are shifted from each other in the transport direction of the original. In FIG. 2, as an example, the reading positions of the image reading units 8 and 9 are arranged point-symmetrically in the cross section of FIG. 2, and the reading positions are shifted upstream and downstream.

In this embodiment, basically, it is assumed that when a document is conveyed, the front side is conveyed downward and the image is read. Described as reading. The same is true for the carrier sheet sandwiching the document.

Further, as shown in FIG. 2, when the control unit 102 receives a document reading command from an information processing apparatus such as a PC, the feed roller 4 and the separation pad 5 feed the document in the paper feed tray 2 one by one through the transport path. separated and fed. Subsequently, the document is transported toward the image reading units 8 and 9 by the transport roller pair 6 . Although not shown, a registration sensor is provided between the transport roller pair 6 and the image reading units 8 and 9 to detect the arrival and passage of the document. Then, when the arrival of the document is detected by the registration sensor, the control section starts reading when the document reaches the reading start position P of the image set at the position on the upstream side of the image reading sections 8 and 9. It controls the image reading units 8 and 9.

Specifically, when the document arrives at the image reading start position, the light emitting section emits R (red) light, and the light receiving section receives the light reflected from the document. Subsequently, the light irradiating section emits G (green) light, and the light receiving section receives the light reflected by the document. Subsequently, B (blue) light is emitted by the light emitting unit and received by the light receiving unit. An image of one line can be read by synthesizing these three light receiving results. The control unit 102 transfers the RGB image data for one line to the image processing unit 103 (see FIG. 3). During this time, the pair of conveying rollers 6 and 7 are always rotated to convey the document, and the image reading units 8 and 9 continue to read the image until the trailing edge of the document passes. The passage of the trailing edge of the document is determined by the elapse of a predetermined time after the trailing edge of the document is detected by the registration sensor described above, or by other document detection sensors provided downstream of the image reading units 8 and 9. It can be detected by detecting the original. The image reading apparatus 100 repeats the above-described operations to read a document in color and output image data.

Next, means for detecting the document area will be described with reference to the flow chart shown in FIG.

For the line data of the entire surface image (A in FIG. 1) input to the image processing unit 103, the left end of the document area is scanned by scanning the line data from the left side, and the document area is scanned by scanning the line data from the right side. Find the right edge of .

First, processing for detecting the left edge of the document area is performed. In this case, when the background plate at the time of reading the document is black or has a brightness close to it, the detection process is performed using the fact that the brightness of the pixels in the document area is high. A description will be given of a situation in which the portion of the carrier sheet read from the surface is also of high brightness. Let B(x) be the brightness of the x-th pixel (the number of pixels from the left end of the image data) of the line data in the image data of the surface, and let Th ₁ be the predetermined threshold. At this time, it is determined whether or not the following expression 1 is satisfied (step S201).

B(x) < Th ₁ (Formula 1)

If Equation 1 holds true in step S201, the pixel of interest is determined to be outside the document area (step S203). Next, it is determined whether the target pixel is the last pixel of the line data, ie, the right end (step S204). If it is determined in step S204 that the end of the line data has been reached, the process ends. If it is determined in step S204 that it is not the end pixel of the line data, the pixel of interest is shifted to the next pixel on the right (by incrementing x) (step S205), and determination is continued.

The determination process described above is continued until the right end of the line data is reached, or until it is determined in step S201 that equation 1 no longer holds, that is, it is determined to be within the document area.

If Equation 1 does not hold in step S201, it is determined that the pixel of interest is within the document area, and the current coordinates are stored as the left edge a of the document area (step S202), and the process ends.

If the pixel of interest is determined to be within the document area in step S201, or if it is determined to be the end pixel in step S204, the process proceeds to the next line and restarts the processing of this flowchart.

Once the left edge a of the document area has been detected, the same process is performed from the right side to detect the right edge of the document area. First, it is checked whether Expression 1 is satisfied (step S201), in the same manner as the above-described processing when scanning from the left side. Subsequent processing may be performed by executing processing equivalent to the processing in which "left" and "right" are interchanged in the above description, and storing the number of pixels from the right end of the detected image data as the end portion b'. . However, in the determination of step S204, it is determined whether it is the edge a stored in the left edge detection processing, not whether it is the end pixel (S204). If the target pixel has been scanned to the left edge a, it is determined that this line is not in the document area, the edge a is deleted, and the left edge detection processing is executed again from the next line.

The number of lines when the left edge a and the edge b' are detected is defined as the upper edge t of the document. Also, the right edge b of the document as the coordinates is calculated by setting the right edge b=T−b′ with respect to the total number of pixels T in the horizontal direction of the image.

By repeating the above from the top of the image, the left edge a, right edge b, and top edge t of the document area in the image can be detected. In the present embodiment, since processing can be performed without using the bottom edge of the document in the processing described later, the bottom edge is not detected. You can use it.

Next, a method for detecting diagonal lines written on the carrier sheet from the back image will be described with reference to FIG.

Using the t-th line at the upper end of the document detected for the front side image as a reference, for the line data of the y-th line from the t-th line, the pixel of interest in the line data of the back side image (B in FIG. 1) is set from the left end a of the document area. Diagonals are detected by scanning to the right. The initial value of y is 0. In reality, the coordinates of the left edge a and the right edge b should be symmetrical with respect to the center in the main scanning direction due to the symmetry of the document transport position in the width direction between the front side and the back side. , for the sake of simplification, common symbols are used for the front surface and the back surface.

First, processing is performed to detect whether the pixel of interest is a diagonal line. Here, the case where the main color of the back surface of the carrier sheet is white or a color close to it, and the diagonal line is drawn in black or a color close to it will be described _. and At this time, it is determined whether or not the following expression 2 is satisfied (step S301). Note that x means the number of pixels from the left end a of the pixel of interest.

B(x) > Th ₂ (Formula 2)

In step S301, it is determined whether or not Equation 2 holds. If Equation 2 holds, it is determined that the pixel of interest is not a diagonal line (step S303), and the process proceeds to step S304.

In step S304, it is determined whether or not the target pixel is equal to or greater than the right end b, which is the end of the processing here. If it is determined that the pixel of interest is not at or above the right edge b of the document area, the pixel to the right of the pixel of interest is referred to (by incrementing x) to continue the determination (step S305). If it is determined in step S304 that the pixel of interest is greater than or equal to the right end b, it means that a diagonal line cannot be found in that line. do.

If Equation 2 does not hold in step S301, the pixel of interest is determined to be the pixel on the left end of the diagonal line, the current coordinates are stored as the coordinates of the diagonal line _Ly (step S302), and the process ends.

The determination process described above is repeated until the bottom edge of the image is reached or, if the bottom edge t' of the original image has been calculated in the same way as the top edge t, it is determined that scanning up to that line has been completed.

Next, the process of correcting the stretch of the image from the diagonal coordinate data will be described with reference to FIG. First, at a value Δ indicating the number of lines determined in advance from the diagonal angle of the carrier sheet, a reference amount of change σ is obtained as follows (S401). The reference amount of change .sigma. is an integer and represents the number of pixels indicating how many pixels the diagonal line changes in the main scanning direction when advancing .DELTA. lines. Further, in the above explanation of the processing, it is assumed that the diagonal line has a width of a plurality of pixels, and the pixel on the left end is detected, and the line connecting them is treated as a diagonal line. For the sake of simplification, the line connecting the left ends of the actual diagonal lines is sometimes referred to as a diagonal line.

σ = L _{t + Δ} - L _t (Formula 3)

Further, the amount of change σ _y for each y is obtained as follows (S402).

_σy =Ly _+Δ − _Ly (Formula 4)

Next, σ _y is compared with σ to confirm whether σ _y is smaller (S403). If _σy is smaller, it is determined that the image is stretched, and the stretch amount is corrected. First, Δ' that satisfies the following equation 5 is obtained (S404).

σ = L _{y + Δ'} - L _y (Formula 5)

Then, reduction processing is performed on the same coordinates of the back side image and the front side image so that the line data of the Δ' line from the y-th line becomes the Δ line (S405). As an example of reduction processing, thinning processing may be performed so that the intervals become uniform.

If it is determined in S403 that _σy is not smaller, it is determined that there is no image stretching, and the process advances to S406 and S407 to determine the next line. If it is the end line in S407, the process ends.

By performing the above processing up to the lower end of the image or document area, the image elongation can be corrected.

Here, FIGS. 7 and 8 show an example of the back side image when performing the process of correcting the elongation of the image as described above. In the following description, only the back side image is subjected to reduction processing, but in the present invention, the same processing is performed to the position of the front side image corresponding to the position of the back side image. However, the description is simply omitted for the sake of simplification.

7 and 8, the vertical axis indicates the number of lines, and the horizontal axis indicates the number of pixels in the main scanning direction. The left side of FIG. 7 shows an example of the reference change amount σ. In this example, the reference change amount σ is a shift of 5 pixels in 5 lines from the 4th line to the 9th line. The right side of FIG. 7 and FIG. 8 show diagrams in the order of processing.

First, the diagram on the upper right side of FIG. 7 shows a state in the middle of processing, and shows a state in which y in S402 of FIG. 6 is set to 3. At this time, Equation 4 becomes σ _y =L _{y +Δ} −L _y =5, and since σ _y =σ, y is incremented by one.

As shown in the middle diagram on the right side of FIG. 7, when y=4, Equation 4 is σ _y =L _y+Δ −L _y =4. At this time, σ _y <σ, so S403 of FIG. 6 becomes Yes, and the process proceeds to S404.

The lower right diagram in FIG. 7 shows a state in which Δ is increased by one to obtain Δ′ that satisfies Equation 5. Equation 4 in this state is σ _y =L _y+Δ −L _y = Since it is 4, Δ is further increased by one.

The upper diagram of FIG. 8 shows a state in which delta is further increased by one, and Equation 4 at this time is σ _y = Ly _{+ Δ} - _Ly = 5, and σ _y = σ there is Proceeding to S405 in FIG. 6, reduction processing is executed.

In the example shown in the middle diagram of FIG. 8, reduction processing is executed by thinning out the 9th and 10th lines for which the value of _σy did not change when increasing the delta in the processing of S403 in FIG. are doing. This is just an example, and reduction processing may be performed by processing such as averaging the pixel values of lines in which _σy does not change (8th to 10th lines).

As a result of the reduction processing, _σy up to the 4th line matches σ, so the process proceeds to S406 in FIG. 6 to shift the target line and repeat the processing of S402 from the 5th line.

By repeating the processing as described above, the elongation of the document image can be corrected using the back side image including the diagonal line.

In addition, in the comparison processing of the amount of change σ _y described above, the comparison is performed for each y, but the comparison processing may be performed at predetermined intervals of two or more lines. In that case, the reduction processing may be performed in units of a predetermined interval as it is during the reduction processing, or only the processing in steps S404 and S405 may be performed for each line as described above. According to this, the comparison process itself can be executed at predetermined intervals to shorten the processing time, and the reduction process can be applied in units of lines.

Further, in the reduction processing described above, when the amount of change σ _y cannot be matched with σ by the thinning processing in units of one line, the reduction processing may be performed collectively in units of a plurality of lines.

Also, in the processing for correcting the elongation of the image described above, an example has been described in which the amount of change σ is treated as an integer indicating how many pixels have changed, but the present invention is not limited to this. For example, the number of pixels changed when advancing Δ lines may be treated as the number of pixels up to about the first decimal place as the slope _σy . It is also possible to calculate all the lines that do not match σ and use the magnitude relation with respect to σ to perform scaling processing so that the gradient _σy between the lines matches or approaches σ as much as possible.

In the above-described processing, the entire image including the carrier sheet is processed. However, the document image area is determined from the front surface image, and a predetermined range in which the document image area or a margin area is provided around the document image area is determined. A series of processes may be applied only to In that case, the mark M may not be found in the back side image, but in that case, additional determination may be made such as not applying the enlargement/reduction process.

In this embodiment, an example in which only one diagonal line is provided on the back surface of the carrier sheet has been described, but two or more diagonal lines may be provided in the same manner.

Also, instead of a diagonal line, a slanted line may be used so that the elongation can be determined based on the angle. For example, two non-perpendicular and non-parallel oblique lines may be drawn on the back surface of the carrier sheet.

In any case, the extension of the document image is corrected by using the linear mark M, which is included in the back side image and whose coordinate position in the main scanning direction changes continuously at the same inclination with respect to the position in the sub-scanning direction. Good to have. Note that if the document is read obliquely, for example, it may not be possible to detect the feeding speed of the document (image stretch) using the angle of the diagonal line. By using one of them, it is possible to detect and correct the conveying speed of the document by using the angle. As a straight line whose coordinate position in the main scanning direction changes continuously at the same inclination with respect to the position in the sub-scanning direction, when the document is skewed, the shadow corresponding to the frame of the document itself and the brightness of the reading background If the frame can be detected by the difference, the frame line may be used.

In the present embodiment, thinning processing is performed when performing image reduction processing, but other reduction processing may be performed. For example, reduction processing may be performed by averaging each line data.

In this embodiment, when the elongation of the back side image is detected, the reduction process is performed for the same coordinates on the front side. Reduction processing may be performed. By doing so, even if the positions of the reading sensor on the front side and the reading sensor on the back side are misaligned, the stretched position of the front side image can be accurately corrected. Alternatively, the correction process may be performed only on the front side, and the back side image may be discarded after the elongation is detected.

In this embodiment, only reduction processing is performed in response to detection of image expansion, but conversely, enlargement processing may be performed in response to detection of image shrinkage. When the enlargement process is performed, Δ' is obtained so that _{δy and δ} are the same even when δy and δ are compared in S403 and δy is large. After that, the data for Δ lines may be output from the data for Δ' lines by interpolation processing. By doing so, for example, it is possible to cope with image shrinkage that occurs when the transport speed becomes faster than the original due to the trailing end of the original slipping off from the place where the transport resistance was applied by the separating portion or the like.

As described above, according to the present invention, a linear mark M provided on the back surface of a carrier sheet and whose coordinate position in the main scanning direction changes continuously at the same inclination with respect to the sub-scanning direction is read, and the mark M is read. It is possible to detect the expansion/contraction of the original image based on the result of reading, and perform enlargement/reduction processing (enlargement or reduction processing) on the image of the original, thereby correcting the image to an appropriate magnification.

In the above-described embodiment, the marks M provided on the back surface of the carrier sheet are linearly changed in position in the main scanning direction (width direction) with the same inclination with respect to the sub-scanning direction (conveyance direction). However, if the shape of the mark M is input to the image processing unit 103 in advance, it may be a logarithmic change, an exponential change, or a monotonically increasing or monotonically decreasing curve. Furthermore, as long as the shape (type) of the mark M is designated to the image processing unit 103 in advance, the mark M does not necessarily have to be linear, and may have some pattern. The image processing unit 103 can detect expansion or contraction of the image in a non-matching portion by performing pattern matching with the shape of the mark M input in advance. The shape (kind) of the mark M may be determined based on the results of reading other marks provided on the carrier sheet itself.

8, 9 image reading unit 100 image reading device 102 control unit 103 image processing unit 104 image output unit M mark

Claims (3)

In an image reading system having an image reading device that reads an image of a document through a surface of a carrier sheet conveyed with the document sandwiched therebetween,
reading a linear mark provided on the back surface of the carrier sheet and having coordinate positions in the width direction perpendicular to the conveying direction continuously changing at the same inclination with respect to the conveying direction in which the carrier sheet is conveyed; An image reading system, wherein an image of the document is enlarged or reduced based on the reading result. Detecting expansion/contraction of the back image according to the reading result of the mark, and performing the enlargement/reduction processing on the front image corresponding to the position in the conveying direction in the back image where the expansion/contraction is detected. 2. The image reading system according to claim 1, characterized by: 3. The image reading system according to claim 1, wherein the marks are provided so as to form diagonal lines on the back surface of the carrier sheet.

Images