CN116862815A - Image sensor seam correction method, system, electronic device and storage medium - Google Patents
Image sensor seam correction method, system, electronic device and storage medium Download PDFInfo
- Publication number
- CN116862815A CN116862815A CN202311133832.6A CN202311133832A CN116862815A CN 116862815 A CN116862815 A CN 116862815A CN 202311133832 A CN202311133832 A CN 202311133832A CN 116862815 A CN116862815 A CN 116862815A
- Authority
- CN
- China
- Prior art keywords
- fusion
- image
- correction
- correction coefficient
- coefficient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
本发明提出一种图像传感器拼缝校正方法、系统、电子装置及存储介质。该方法包括：在均匀光源下，采集若干不同亮度的原始图像，记录原始图像的像素值；基于所述像素值，计算原始图像的校正系数；分割所述原始图像，使得有效子区域内含有拼缝线及拼缝线两侧的传感器图像；基于任一有效子区域图像的像素值，计算该有效子区域对应的区域校正系数；构建融合系数函数，基于校正系数和区域校正系数，计算融合校正系数；基于融合校正系数，校正所述像素值，输出校正图像。本发明提出一种分段融合校正方法，在物理空间不同区域采用不同的校正系数，降低成像拼缝效应，使得该方法根据环境变化重新校正系数，具有自适应能力。
The invention proposes an image sensor seam correction method, system, electronic device and storage medium. The method includes: collecting several original images of different brightnesses under a uniform light source, recording the pixel values of the original images; calculating the correction coefficient of the original images based on the pixel values; segmenting the original images so that the effective sub-regions contain spelling Sensor images on both sides of the suture and stitching line; based on the pixel value of any effective sub-region image, calculate the regional correction coefficient corresponding to the effective sub-region; construct a fusion coefficient function, and calculate the fusion correction based on the correction coefficient and the regional correction coefficient Coefficient; based on the fusion correction coefficient, correct the pixel value and output the corrected image. The present invention proposes a segmented fusion correction method, which uses different correction coefficients in different areas of the physical space to reduce the imaging patchwork effect, so that the method re-corrects the coefficients according to environmental changes and has adaptive capabilities.
Description
Technical Field
The present invention relates to the field of image processing, and in particular, to a method, a system, an electronic device, and a storage medium for correcting a seam of an image sensor.
Background
Existing image sensor non-uniformity correction methods mainly include dark field non-uniformity correction and light response non-uniformity correction, where the corrected image sensor is typically a full-film image sensor, rather than a tiled image sensor. As shown in fig. 1, in order to increase the resolution of the sensor, the large area sensor adopts a plurality of sensors for stitching, and the stitching can form obvious vertical or horizontal stitching lines (stitching lines) on the photographed image, so that the imaging effect is affected. The imaging of the physical stitching line is related to the angle of the light source and the lens, the fixed correction coefficient cannot adapt to the image acquisition environment and the change of the lens, and the traditional transverse and vertical stripe correction algorithm cannot completely eliminate the stitching effect.
Chinese patent CN112291446B discloses a method for correcting non-uniformity of a large area array CMOS image sensor. Firstly, under uniform illumination of an integrating sphere, original image acquisition is carried out on a CMOS image sensor, non-uniformity calculation is carried out on the acquired image, an image corresponding to the maximum non-uniformity value is obtained, then, according to the structural characteristics of the CMOS image sensor, the array of the large-area array CMOS image is subjected to grouping processing, parameter calculation is carried out by grouping, and finally, a correction model for carrying out non-uniformity correction on the CMOS image sensor is established. This patent differs from the solution of the present invention.
The invention provides a method and a device for correcting the non-uniformity of the light response on a COMS image sensor, which can calculate the offset of an image under different saturation levels off line in the process of correcting the non-uniformity, only need to know a standard correction matrix in the process of on-line calculation, obtain the correction matrix parameters under the saturation level through the offset under different saturation levels, introduce a linear interpolation method, a nonlinear interpolation method and a nearest neighbor principle in the process of correcting calculation, calibrate the standard parameter matrix and increase the correction precision. This patent differs from the solution of the present invention.
Chinese patent CN 102176742B proposes an image correction coefficient acquisition method, a non-uniform image correction method and a system, wherein at least two groups of sensor units in a sensor array are adopted, a plurality of background images under different working temperature conditions of the sensor array are acquired, pixels of a pre-corrected image output by the sensor array are correspondingly grouped according to the grouping of the sensor units in the sensor array, pixel values of the pre-corrected image are acquired according to the groups, an average value of the pixel values in each group is calculated, and in the grouping of the pre-corrected image, the correction coefficient of the current background is calculated under the condition that the square sum of the average value difference of every two groups of pixel values of the corrected image is minimum. This patent is different from the technical problem solved by the present invention.
Chinese patent CN113450270a discloses a correction parameter generating method, an electronic device and a storage medium, which acquire a digital image of a single-color object under uniform illumination condition through an image sensor to be corrected; determining a photosensitive response uneven baseline in the digital image, and determining a first image area and a second image area which are symmetrical to each other at two sides of the photosensitive response uneven baseline; initializing a first correction parameter of each first pixel in the first image area; determining a second correction parameter which is symmetrical to each second pixel in a second image area of each first pixel by taking a photosensitive response uneven baseline as a symmetry axis by taking each first pixel in the first image area as a reference standard; according to the first correction parameter and the second correction parameter, the correction parameter of the image sensor to be corrected is determined, the problem that the image sensor in the related technology has photosensitive response non-uniformity is solved, and the photosensitive response uniformity of the image sensor is improved. This patent is different from the technical problem solved by the present invention.
Disclosure of Invention
The invention provides a method, a system, an electronic device and a storage medium for correcting an image sensor seam, which can at least solve one of the technical problems.
In order to achieve the above purpose, the present invention proposes the following technical solutions:
an image sensor seam correction method, comprising:
under a uniform light source, collecting a plurality of original images with different brightness, and recording pixel values of the original images;
calculating a correction coefficient of the original image based on the pixel value;
dividing the original image to enable the effective region to contain a stitching line and sensor images on two sides of the stitching line;
calculating a region correction coefficient corresponding to an effective sub-region based on a pixel value of any effective sub-region image;
constructing a fusion coefficient function, and calculating a fusion correction coefficient based on the correction coefficient and the regional correction coefficient;
and correcting the pixel value based on the fusion correction coefficient, and outputting a corrected image.
Further, the constructing a fusion coefficient function includes: setting left fusion width L in sensor images on the left and right sides of the stitching line position S as an axis 1 Right fusion width L 2 The method comprises the steps of carrying out a first treatment on the surface of the And constructing a fusion coefficient function f (i) around the stitching line position S within a set left and right fusion width range, wherein i represents the number of columns or rows of image pixels.
Further, the fusion coefficient function f (i) satisfies f (S-L 1 )=0，f(S+L 2 ) =0, f (S) =1, and f (i) is at [ S-L 1 ,S]Monotonically increasing in range, f (i) is at [ S, S+L 2 ]Monotonically decreasing in range.
Further, the sum of fusion widths set in any one of the sensor images is smaller than the width of the sensor image.
Further, the correcting the pixel value based on the fusion correction coefficient, before further includes: and storing the fusion correction coefficient corresponding to any effective subarea.
Further, the correcting the pixel value based on the fusion correction coefficient includes: and finding out the effective subarea based on the pixel value and the pixel position of the original image, determining a fusion correction coefficient corresponding to the effective subarea, correcting the pixel value, and outputting a corrected image.
The invention also provides an image sensor seam correction system, which comprises:
the image acquisition unit acquires a plurality of original images with different brightness under a uniform light source and records pixel values of the original images;
an image segmentation unit for segmenting the original image so that the effective region contains a stitching line and sensor images on two sides of the stitching line;
a parameter calculation unit that calculates a correction coefficient of an original image based on pixel values of the original image; calculating a region correction coefficient corresponding to an effective sub-region based on a pixel value of any effective sub-region image; constructing a fusion coefficient function, and calculating a fusion correction coefficient based on the correction coefficient and the regional correction coefficient;
and the correction unit is used for finding out the effective subarea based on the pixel value and the pixel position of the original image, determining a fusion correction coefficient corresponding to the effective subarea, correcting the pixel value and outputting a corrected image.
Further, the method further comprises the following steps:
and the storage unit is used for storing the fusion correction coefficient corresponding to any effective subarea.
The invention also proposes an electronic device comprising a memory in which a computer program is stored and a processor arranged to run the computer program to perform the image sensor seam correction method as described above.
The present invention also proposes a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the image sensor seam correction method as described above.
The beneficial effects of the invention are as follows: the invention provides a segmentation fusion correction method, which adopts different correction coefficients in different areas of a physical space to reduce an imaging seam effect, so that the method can recalibrate the coefficients according to environmental changes and has self-adaption capability.
Drawings
FIG. 1 is a schematic diagram of an image sensor stitching effect;
FIG. 2 is a flowchart of a method for correcting an image sensor seam in embodiment 1 of the present invention;
FIG. 3 is a schematic diagram of a split segment correction according to embodiment 1 of the present invention;
FIG. 4 is a schematic diagram of a multi-sensor image setting fusion width in embodiment 1 of the present invention;
FIG. 5 is an image of the fusion coefficient function in example 1 of the present invention;
FIG. 6 is a flowchart of image correction in embodiment 1 of the present invention;
FIG. 7 is an image before correction of a seam segmentation in example 1 of the present invention;
FIG. 8 is an image after the patch correction in embodiment 1 of the present invention;
FIG. 9 is a schematic diagram of a split segment correction according to embodiment 2 of the present invention;
FIG. 10 is an image of the fusion coefficient function in embodiment 2 of the present invention;
FIG. 11 is an image of the fusion coefficient function in embodiment 3 of the present invention;
fig. 12 is an image of the fusion coefficient function in embodiment 4 of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention.
A horizontal physical stitching line will form a horizontal demarcation line on the image and a vertical physical stitching line will form a vertical demarcation line on the image, and a similar correction algorithm is called either cross-bar correction or vertical-bar correction. The correction thought of the horizontal stitching line is identical to that of the vertical stitching line, and only a vertical dividing line (vertical correction) is taken as an example for illustration in the embodiment of the invention. The correction method provided by the invention is also suitable for the correction of the seam in the horizontal direction, and the correction of the seam in the horizontal direction can be carried out only by exchanging row and column coordinates in the scheme.
Example 1
As shown in fig. 2, the embodiment takes a method of correcting a seam in a vertical direction as an example, and specifically includes:
under the uniform light source, a plurality of original images with different brightness are collected, and pixel values of the original images are recorded, wherein the pixel values comprise row pixel values and column pixel values.
The invention is described by taking a vertical seam correction method as an example, so that the pixel values recorded in the invention are recorded as column pixel values by taking the column coordinates of the image as references. The original image includes several sensor images with a stitching line as a boundary of the sensor images.
In addition, if the horizontal stitching line is corrected, the line coordinate is required to be used as an index, and the pixel value of the original image is recorded as a line pixel value; the specific correction method is the same as the correction method of the vertical seam line, and only column coordinates in the correction method of the seam line in the vertical direction are required to be replaced by row coordinates to be corrected.
In this embodiment, several images with different brightness are collected under a uniform light source, and any pixel (column pixel value) x in the vertical direction of any image is recorded m,i Wherein the subscript m represents the number of steps of the current light source brightness and i represents the ith column of the image.
At the brightness of the first-level light source, the column pixel value x of the original image is recorded 1,i Is denoted as (x) 1,1 、x 1,2 、...x 1,i ) The method comprises the steps of carrying out a first treatment on the surface of the Converting light source brightness, recording column pixel value x of original image under second-level light source brightness 2,i Is denoted as (x) 2,1 、x 2,2 、...x 2,i )。
Based on the pixel values, correction coefficients of the original image are calculated.
In this embodiment, taking the pixel value of the original image column under the two-stage light source brightness as an example, a traditional vertical streak correction algorithm is adopted to calculate the correction coefficient of the original image. Wherein the correction coefficient of the original image comprises a multiplication coefficient k i And addition coefficient b i 。
Respectively calculating column pixel mean value y according to the recorded column pixel values of the original images with different brightness m,i According to any column pixel mean y m,i Calculate the corresponding correction coefficient, including multiplication coefficient k i And addition coefficient b i The calculation formula is as follows:
traditional vertical line schoolIn the positive algorithm, the correction multiplication coefficient k is obtained according to calculation i And addition coefficient b i For any pixel p r,i And correcting, wherein the correction formula is as follows: p is p 0 =(p r,i +b i )×k i . Wherein the subscript r represents the image row r and i represents the image column i; p is p r,i Pixel values representing the ith row and ith column of any sensor image are also pixel input values; p is p 0 Output values for the pixels.
Preferably, in practical application, a plurality of (at least two) light source brightnesses can be set, and column pixel values of corresponding sensor images are acquired. If the pixel values of the sensor image column under the brightness of multiple (more than two) light sources are recorded, other calculation modes such as a least square method can be used to calculate the average value of the column pixels.
The original image is segmented so that the effective region contains stitching lines and sensor images on both sides of the stitching lines.
The number and the direction of the dividing lines can be set freely, a plurality of effective subareas are needed in the dividing result, any effective subarea contains the stitching line and the sensor images at two sides of the stitching line, and all the effective subareas can restore the stitching line and the sensor images at two sides of the stitching line. In the present invention, at least one dividing line is required to divide the entire image into 2 areas.
As shown in fig. 3, in this embodiment, a dividing line is provided along a direction perpendicular to the stitching line, dividing the original image into upper and lower horizontal regions. When the seam is along the vertical direction, the number of the dividing lines is less than half of the number of lines of the sensor image; when the seam is along the horizontal direction, the number of the dividing lines is less than half of the number of the sensor image columns.
And calculating a region correction coefficient corresponding to the effective sub-region based on the pixel value of any effective sub-region image.
In this embodiment, referring to fig. 3, the original image is divided into two upper and lower horizontal areas, and any horizontal area contains a stitching line and two side sensor images, so that the horizontal areas all belong to the effective sub-areas.
Calculating region corrections for each horizontal region jThe specific process is as follows: according to the recorded column pixel value of the original image, finding the column pixel value x corresponding to the horizontal region j m,i,j Wherein the subscript m represents the number of steps of the current light source brightness, i represents the ith column of the image, and j represents the region number.
At the first level of light source brightness, the column pixel value corresponding to the horizontal region j is denoted as (x) 1,1,j 、x 1,2,j 、...x 1,i,j ) The method comprises the steps of carrying out a first treatment on the surface of the The column pixel value corresponding to the horizontal region j is denoted as (x) at the second level light source luminance 2,1,j 、x 2,2,j 、...x 2,i,j )。
Respectively calculating regional column pixel mean y j,i,m And calculates the corresponding region multiplication coefficient k j,i Sum-region addition coefficient b j,i The calculation formula is as follows:
and constructing a fusion coefficient function, and calculating a fusion correction coefficient based on the correction coefficient and the regional correction coefficient.
The method for constructing the fusion coefficient function specifically comprises the following steps: setting left fusion width L in sensor images on the left and right sides of the stitching line position S as an axis 1 Right fusion width L 2 So that the total fusion width is 2L; within the total fusion width, a fusion coefficient function f (i) is constructed around the stitching line position S, where i represents the number of columns or rows of image pixels. Total fusion width 2l=l set in the present invention 1 +L 2 。
The fusion coefficient function satisfies f (S-L 1 )=0，f(S+L 2 ) =0, f (S) =1, and f (i) is at [ S-L 1 ,S]Monotonically increasing in range, f (i) is at [ S, S+L 2 ]Monotonically decreasing in range.
The range of the total fusion width is related to the number of rows and columns of the sensor and the positions of the seams, and the sum of the fusion widths set in any sensor image is smaller than the width of the sensor image.
As shown in fig. 4, the original image is stitched from 4 sensor images, containing 3 physical stitching lines.The width of the sensor image 1 is d 1 I.e. the distance d between the physical splice line 1 and the left boundary 1 The method comprises the steps of carrying out a first treatment on the surface of the The width of the sensor image 2 is d 2 Also the distance d between the physical splice line 1 and the physical splice line 2 2 The method comprises the steps of carrying out a first treatment on the surface of the The width of the sensor image 3 is d 3 I.e. the distance d between the physical splice line 2 and the physical splice line 3 3 The method comprises the steps of carrying out a first treatment on the surface of the The width of the sensor image 4 is d 4 I.e. the distance between the physical splice line 3 and the right boundary.
The total fusion width range is set for the physical splice line 1, and specifically comprises the following steps: setting a left fusion width L in the sensor image 1 1 The right fusion width L is set in the sensor image 2 2 . The total fusion width range is set for the physical splice line 2, and specifically comprises the following steps: setting a left fusion width L in the sensor image 2 3 A right fusion width L is set in the sensor image 3 4 . The total fusion width range is set for the physical splice line 3, and specifically comprises: setting a left fusion width L in the sensor image 3 5 The right fusion width L is set in the sensor image 4 6 。
Wherein the fusion width L set in the sensor image 1 1 <d 1 The method comprises the steps of carrying out a first treatment on the surface of the Fusion width L set in sensor image 2 2 +L 3 <d 2 The method comprises the steps of carrying out a first treatment on the surface of the Fusion width L set in sensor image 3 4 +L 5 <d 3 The method comprises the steps of carrying out a first treatment on the surface of the Fusion width L set in sensor image 4 6 <d 4 。
As shown in fig. 5, the present embodiment proposes that the function f (i) satisfying the above requirements is defined as follows:
the function f (i) proposed in the present embodiment is a linear function, and is only used as an example of a fusion coefficient function, and the specific definition of the fusion coefficient function may be set according to the actual situation and the requirement.
And calculating a fusion correction coefficient based on the determined fusion coefficient function, the correction coefficient of the original image and the region correction coefficient. For the horizontal region j, the calculation formula of the fusion correction coefficient is as follows:
in this embodiment, a nonvolatile memory is further provided, and is configured to store the calculated fusion correction coefficient, so that the fusion correction coefficient is conveniently found in a subsequent correction process, and correction efficiency is improved.
And correcting the pixel value based on the fusion correction coefficient, and outputting a corrected image. The method specifically comprises the following steps: and finding out the effective subarea according to the pixel position (r, i), determining a fusion correction coefficient corresponding to the effective subarea, correcting the pixel value, and outputting a corrected image.
As shown in fig. 6, the region j to which the current pixel belongs is calculated according to the pixel position (r, i), and the corresponding fusion correction coefficient is read from the nonvolatile memoryAnd->For pixel value p r,i The linear correction formula is adopted:performing linear correction; outputting the correction result p out Replacement pixel value p r,i The linearity correction process is completed.
Wherein the subscript r represents the image row r and i represents the image column i; p is p r,i Pixel values representing the sensor's r-th row and i-th column, are also pixel input values; p is p out Output values for the pixels.
The pre-processing image of this embodiment is shown in fig. 7, and the processed image is obtained by the correction method described above, see fig. 8.
Preferably, a memory is not set, and a fusion correction coefficient corresponding to any effective sub-region is not stored. At this time, the correction specific process is: calculating the region j of the current pixel through the pixel position (r, i), repeating the calculation process of the parameters, and calculating the corresponding fusion correctionPositive coefficientAnd->For pixel value p r,i The linear correction formula is adopted:performing linear correction; outputting the correction result p out Replacement pixel value p r,i The linearity correction process is completed.
Example 2
On the basis of embodiment 1, a new fusion coefficient function is proposed in this embodiment. The concrete construction process is as follows:
setting the symmetrical ranges of the left side and the right side as fusion width L by taking the stitching line position S as an axis, and setting the total fusion width as 2L; as a fusion coefficient function f (i), an axisymmetric function around the stitching line position S is used within the fusion width range, where i represents the number of columns or rows of image pixels.
The specific setting of the fusion width in this embodiment is shown in fig. 9.
The fusion coefficient function satisfies f (S-L) =0, f (s+l) =0, f (S) =1, and f (i) monotonically increases in the range of [ S-L, S ].
The present embodiment proposes that the function family f (i) satisfying the above requirements be defined as follows:
where the variable n is used to control the trend of the monotonically increasing function over the range S-L, S, L representing half the fusion width. The function family proposed in the embodiment is only an example of a fusion coefficient function, and the specific definition of the fusion coefficient function can be set according to the actual situation and the requirement.
As shown in fig. 9, the seam line is spaced from the left boundary of the left sensor 1 by a distance d 1 The distance between the stitching line and the right boundary of the right sensor 2 is d 2 The fusion width satisfies L<min(d 1 ,d 2 ). In order to better embody the effect of the fusion effect of the region division, the fusion width L is set in this embodiment<min(d 1 /2,d 2 /2)。
In practical application, the corresponding fusion coefficient function f (i) can be calculated by taking n values for a plurality of times, and the final n value is determined according to the fusion effect of f (i), so as to determine the fusion correction coefficient.
In this embodiment, as shown in fig. 10, taking n=1, a fusion coefficient function f (i) is constructed as follows:
example 3
In this embodiment, different fusion coefficient functions f (i) are set on the basis of embodiment 2. When 0< n <1, the function monotonically increases faster and faster within the [ S-L, S ] range. When n=1/2, the image of the function f (i) is shown in fig. 11, and the specific expression is as follows:
example 4
In this embodiment, different fusion coefficient functions f (i) are set on the basis of embodiment 2. When n >1, the function monotonically increases more and more slowly over the [ S-L, S ] range.
When n=2, the function f (i) image is as shown in fig. 12, and the specific expression is as follows:
the invention provides a segmentation fusion correction method, which adopts different correction coefficients in different areas of a physical space to reduce an imaging seam effect, so that the method can recalibrate the coefficients according to environmental changes and has self-adaption capability.
The invention also provides an image sensor seam correction system, which comprises:
the image acquisition unit acquires a plurality of original images with different brightness under a uniform light source and records pixel values of the original images;
an image segmentation unit for segmenting the original image so that the effective region contains a stitching line and sensor images on two sides of the stitching line;
a parameter calculation unit that calculates a correction coefficient of an original image based on pixel values of the original image; calculating a region correction coefficient corresponding to an effective sub-region based on a pixel value of any effective sub-region image; constructing a fusion coefficient function, and calculating a fusion correction coefficient based on the correction coefficient and the regional correction coefficient;
and the correction unit is used for finding out the effective subarea based on the pixel value and the pixel position of the original image, determining a fusion correction coefficient corresponding to the effective subarea, correcting the pixel value and outputting a corrected image.
Further comprises: and the storage unit is used for storing the fusion correction coefficient corresponding to any effective subarea.
The invention also proposes an electronic device comprising a memory and a processor, the memory storing a computer program, the processor being arranged to run the computer program to perform the above-mentioned image sensor seam correction method.
The present invention also proposes a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the image sensor seam correction method as described above.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. An image sensor seam correction method, comprising:
under a uniform light source, collecting a plurality of original images with different brightness, and recording pixel values of the original images;
calculating a correction coefficient of the original image based on the pixel value;
dividing the original image to enable the effective region to contain a stitching line and sensor images on two sides of the stitching line;
calculating a region correction coefficient corresponding to an effective sub-region based on a pixel value of any effective sub-region image;
constructing a fusion coefficient function, and calculating a fusion correction coefficient based on the correction coefficient and the regional correction coefficient;
and correcting the pixel value based on the fusion correction coefficient, and outputting a corrected image.
2. The image sensor seam correction method of claim 1, wherein the constructing a fusion coefficient function comprises: setting left fusion width L in sensor images on the left and right sides of the stitching line position S as an axis 1 Right fusion width L 2 The method comprises the steps of carrying out a first treatment on the surface of the And constructing a fusion coefficient function f (i) around the stitching line position S within a set left and right fusion width range, wherein i represents the number of columns or rows of image pixels.
3. The image sensor seam correction method of claim 2, wherein the fusion coefficient function f (i) satisfies f (S-L 1 )=0，f(S+L 2 ) =0, f (S) =1, and f (i) is at [ S-L 1 ,S]Monotonically increasing in range, f (i) is at [ S, S+L 2 ]Monotonically decreasing in range.
4. The image sensor seam correction method according to claim 2, wherein a sum of fusion widths set in any one of the sensor images is smaller than a width of the sensor image.
5. The image sensor seam correction method according to claim 1, wherein the correcting the pixel value based on the fusion correction coefficient, before further comprising: and storing the fusion correction coefficient corresponding to any effective subarea.
6. The image sensor seam correction method of claim 5, wherein the correcting the pixel values based on the fusion correction coefficient comprises: and finding out the effective subarea based on the pixel value and the pixel position of the original image, determining a fusion correction coefficient corresponding to the effective subarea, correcting the pixel value, and outputting a corrected image.
7. An image sensor seam correction system, comprising:
the image acquisition unit acquires a plurality of original images with different brightness under a uniform light source and records pixel values of the original images;
an image segmentation unit for segmenting the original image so that the effective region contains a stitching line and sensor images on two sides of the stitching line;
a parameter calculation unit that calculates a correction coefficient of an original image based on pixel values of the original image; calculating a region correction coefficient corresponding to an effective sub-region based on a pixel value of any effective sub-region image; constructing a fusion coefficient function, and calculating a fusion correction coefficient based on the correction coefficient and the regional correction coefficient;
and the correction unit is used for finding out the effective subarea based on the pixel value and the pixel position of the original image, determining a fusion correction coefficient corresponding to the effective subarea, correcting the pixel value and outputting a corrected image.
8. The image sensor seam correction system of claim 7, further comprising:
and the storage unit is used for storing the fusion correction coefficient corresponding to any effective subarea.
9. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the image sensor seam correction method of any of claims 1-6.
10. A computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the image sensor seam correction method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311133832.6A CN116862815B (en) | 2023-09-05 | 2023-09-05 | Image sensor seam correction method, system, electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311133832.6A CN116862815B (en) | 2023-09-05 | 2023-09-05 | Image sensor seam correction method, system, electronic device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116862815A true CN116862815A (en) | 2023-10-10 |
| CN116862815B CN116862815B (en) | 2023-11-14 |
Family
ID=88232683
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311133832.6A Active CN116862815B (en) | 2023-09-05 | 2023-09-05 | Image sensor seam correction method, system, electronic device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116862815B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130321678A1 (en) * | 2012-05-31 | 2013-12-05 | Apple Inc. | Systems and methods for lens shading correction |
| CN103856727A (en) * | 2014-03-24 | 2014-06-11 | 北京工业大学 | Multichannel real-time video splicing processing system |
| CN109194872A (en) * | 2018-10-24 | 2019-01-11 | 深圳六滴科技有限公司 | Panoramic image pixel brightness correcting method, device, panorama camera and storage medium |
| US20200182949A1 (en) * | 2018-12-11 | 2020-06-11 | Vulcan Inc. | Magneto optic disk imager |
| CN112085659A (en) * | 2020-09-11 | 2020-12-15 | 中德（珠海）人工智能研究院有限公司 | A panorama stitching fusion method, system and storage medium based on spherical screen camera |
| CN113450270A (en) * | 2021-05-26 | 2021-09-28 | 浙江大华技术股份有限公司 | Correction parameter generation method, electronic device, and storage medium |
| KR20220017697A (en) * | 2020-08-05 | 2022-02-14 | 한국기술교육대학교 산학협력단 | calibration method and apparatus among mutiple sensors |
| CN115423821A (en) * | 2022-08-10 | 2022-12-02 | 中国科学院深圳先进技术研究院 | LED screen splicing area image segmentation method and LED screen bright and dark line correction method |
-
2023
- 2023-09-05 CN CN202311133832.6A patent/CN116862815B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130321678A1 (en) * | 2012-05-31 | 2013-12-05 | Apple Inc. | Systems and methods for lens shading correction |
| CN103856727A (en) * | 2014-03-24 | 2014-06-11 | 北京工业大学 | Multichannel real-time video splicing processing system |
| CN109194872A (en) * | 2018-10-24 | 2019-01-11 | 深圳六滴科技有限公司 | Panoramic image pixel brightness correcting method, device, panorama camera and storage medium |
| US20200182949A1 (en) * | 2018-12-11 | 2020-06-11 | Vulcan Inc. | Magneto optic disk imager |
| KR20220017697A (en) * | 2020-08-05 | 2022-02-14 | 한국기술교육대학교 산학협력단 | calibration method and apparatus among mutiple sensors |
| CN112085659A (en) * | 2020-09-11 | 2020-12-15 | 中德（珠海）人工智能研究院有限公司 | A panorama stitching fusion method, system and storage medium based on spherical screen camera |
| CN113450270A (en) * | 2021-05-26 | 2021-09-28 | 浙江大华技术股份有限公司 | Correction parameter generation method, electronic device, and storage medium |
| CN115423821A (en) * | 2022-08-10 | 2022-12-02 | 中国科学院深圳先进技术研究院 | LED screen splicing area image segmentation method and LED screen bright and dark line correction method |
Non-Patent Citations (3)
| Title |
|---|
| 李勇;杜丙新;: "基于小区域融合的实时视频拼接技术", 吉林大学学报(理学版), no. 06, pages 181 - 186 * |
| 李华, 薛建国: "面阵CCD图像传感器不均匀性的校正", 传感器技术, no. 04, pages 16 - 18 * |
| 董宁 等: "高速率分辨率线性扫描工业相机", 科技成果, pages 1 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116862815B (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111695554B (en) | Text correction method and device, electronic equipment and storage medium | |
| CA2659847C (en) | System and method for adaptive non-uniformity compensation for a focal plane array | |
| WO2022127225A1 (en) | Image stitching method and apparatus, and device and storage medium | |
| CN103530852A (en) | Method for correcting distortion of lens | |
| CN104252700A (en) | Histogram equalization method for infrared image | |
| CN104581098B (en) | A kind of adaptive processing method of lens shading | |
| JP2010252325A (en) | System and method for image correction | |
| CN108550113A (en) | Image scanning output method, device, computer equipment and storage medium | |
| CN118197211B (en) | Data processing methods, apparatus, computer equipment and storage media | |
| CN116862815B (en) | Image sensor seam correction method, system, electronic device and storage medium | |
| CN116563388B (en) | Calibration data acquisition method, device, electronic equipment and storage medium | |
| CN111047521B (en) | Infrared image non-uniformity parameterization correction optimization method based on image entropy | |
| CN114266758A (en) | Panel brightness correction method, device and computer readable storage medium | |
| CN107633487A (en) | A kind of system-level relative radiometric correction method of aviation sweep type multispectral scanner image | |
| CN114007055B (en) | Image sensor lens shading correction method and device | |
| CN112819738B (en) | Infrared image fusion method, device, computer equipment and storage medium | |
| CN114240765B (en) | Interference light removal method, device and storage medium based on grayscale camera | |
| CN117641137A (en) | Method and device for correcting dead pixel of linear array camera | |
| CN115100078A (en) | A method and related device for correcting and filling lattice coordinates in curved screen images | |
| JP5761989B2 (en) | Image recognition apparatus, image recognition method, and program | |
| CN118644412B (en) | A calibration method for image equalization | |
| CN113139912A (en) | Optical fiber image display recovery algorithm and display system | |
| CN117409065B (en) | Object measuring method, device, computer apparatus, readable storage medium and product | |
| CN103929584A (en) | Image correction method and image correction circuit | |
| CN119270630B (en) | Bionic subsoiler control method based on intelligent driving, terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 230000, 3F, Building J2, Phase II, Zhong'an Chuanggu Science and Technology Park, Wangjiang West Road, High tech Zone, Hefei City, Anhui Province Patentee after: Hefei Eko Photoelectric Technology Co.,Ltd. Country or region after: China Address before: 230000 Floor 3, Building J2, Phase II, Zhong'an Chuanggu, High-tech Zone, Hefei City, Anhui Province Patentee before: Hefei Eko Photoelectric Technology Co.,Ltd. Country or region before: China |