CN109360165B - Scene-based blind pixel image restoration method and system - Google Patents
Scene-based blind pixel image restoration method and system Download PDFInfo
- Publication number
- CN109360165B CN109360165B CN201811137990.8A CN201811137990A CN109360165B CN 109360165 B CN109360165 B CN 109360165B CN 201811137990 A CN201811137990 A CN 201811137990A CN 109360165 B CN109360165 B CN 109360165B
- Authority
- CN
- China
- Prior art keywords
- blind
- pixel
- points
- matrix
- line
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 239000011159 matrix material Substances 0.000 claims abstract description 109
- 238000012986 modification Methods 0.000 claims description 7
- 230000004048 modification Effects 0.000 claims description 7
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
Images
Classifications
-
- G06T5/77—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
Landscapes
- Image Processing (AREA)
- Transforming Light Signals Into Electric Signals (AREA)
Abstract
The invention relates to a blind pixel image restoration method and a blind pixel image restoration system based on a scene, wherein a scene image imaged by a thermal infrared imager is obtained; establishing a matrix by taking a blind pixel point in a scene image as a center; determining four directions by taking the blind pixel point as a center, and determining the direction of the blind pixel point according to the ratio of the mean value of the pixel values of the points on the line where the blind pixel point is located to the mean value of the pixel values of the points on the line adjacent to the blind pixel point in the same direction matrix; and modifying and replacing the pixel values of the blind pixel points by the mean value of the pixel values of the points on the line on which the blind pixel points in the matrix with the determined direction consistent with the direction and the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points in the matrix with the same direction are on line. The method carries out restoration and replacement through the directions of the blind pixel points and the pixel value mean values in the adjacent directions so as to correct the known blind pixels of the thermal infrared imager and ensure that the known blind pixels have good consistency with the images formed by the thermal infrared imager.
Description
Technical Field
The invention relates to a method and a system for repairing a blind pixel image, in particular to a method and a system for repairing a blind pixel image based on a scene.
Background
The main component of the infrared thermal imager is an infrared focal plane array, and due to the production process, a certain number of detection elements called blind elements exist, and the blind elements cannot normally work under the condition that the thermal imager is powered on, so that points which cannot truly reflect thermal imaging information exist on an image formed by the infrared thermal imager. In the prior art, for processing a blind pixel point in an image, an averaging method is usually adopted, that is, an average value of points around the blind pixel point is used for replacement, but the directionality of the blind pixel in a scene image is not considered in the method, so that the replacement with the area average value is not a technical problem of replacing the optimal pixel value of the blind pixel and does not have good consistency.
Disclosure of Invention
The technical problem solved by the invention is as follows: a scene-based blind pixel image restoration method and a scene-based blind pixel image restoration system are constructed, and the technical problem that the restored blind pixel images in the prior art are not good in consistency is solved.
The technical scheme of the invention is as follows: the method for repairing the blind pixel image based on the scene comprises the following steps:
acquiring a scene image: acquiring a scene image imaged by a thermal infrared imager;
establishing a matrix: establishing a matrix by taking a blind pixel point in a scene image as a center;
determining the directionality of the blind spot: four directions are determined with the dummy point as the center, namely: determining the direction of the blind element point according to the ratio of the mean value of the pixel values of the points on the line where the blind element point is located to the mean value of the pixel values of the points on the line adjacent to the line where the blind element point is located in the same direction matrix, and determining the direction of the blind element point as the direction of the blind element point according to the consistent direction if the direction of the blind element point is determined to be consistent with any one direction of the four directions;
pixel value replacement: and modifying and replacing the pixel values of the blind pixel points by the mean value of the pixel values of the points on the line on which the blind pixel points in the matrix with the determined direction consistent with the direction and the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points in the matrix with the same direction are on line.
The further technical scheme of the invention is as follows: and when the modification and replacement of the pixel values of the blind pixel point images are determined, performing the modification and replacement of the pixel values of the blind pixel point images by using the mean value of the pixel values of the points on the lines where the blind pixel points in the matrixes in the same direction are located and the mean value of the pixel values of the points on two adjacent lines where the blind pixel points in the matrixes in the same direction are located.
The further technical scheme of the invention is as follows: when the modified and replaced blind pixel point pixel values are calculated, the average ratio of the pixel values of the points on the lines where the blind pixel points are located in the matrix in the same direction is 50%, and the average values of the pixel values of the points on two adjacent lines where the blind pixel points are located in the matrix in the same direction are respectively 25%.
The further technical scheme of the invention is as follows: the condition for determining that the direction of the dummy point coincides with any one of the four directions is: the ratio of the mean value of the pixel values of the points on the line on which the blind pixel points are located to the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points are located in the same-direction matrix is greater than 0.75 and less than 1.25.
The further technical scheme of the invention is as follows: the condition for determining that the direction of the dummy point coincides with any one of the four directions is: the ratio of the mean value of the pixel values of the points on the line on which the blind pixel points are located to the mean value of the pixel values of the points on two lines adjacent to the line on which the blind pixel points are located in the same direction matrix is greater than 0.75 and less than 1.25.
The further technical scheme of the invention is as follows: if there are other blind primitive points in the point on the line where the blind primitive point is located, the pixel value of the other blind primitive points is replaced with the pixel value of any non-blind primitive point on the line.
The further technical scheme of the invention is as follows: the matrix is a 7X7 matrix centered on a blind pixel point in the scene image.
The technical scheme of the invention is as follows: the method comprises the steps of constructing a blind pixel image restoration system based on a scene, wherein the system comprises a scene image acquisition module, a matrix establishment module, a direction determination module for determining the directionality of blind pixel points and a replacement module for replacing pixel values, wherein the scene image acquisition module acquires a scene image imaged by a thermal infrared imager; the matrix establishing module establishes a matrix by taking a blind pixel point in a scene image as a center; the direction determination module determines four directions by taking a dummy point as a center, namely: determining the direction of the blind element point according to the ratio of the mean value of the pixel values of the points on the line where the blind element point is located to the mean value of the pixel values of the points on the line adjacent to the line where the blind element point is located in the same direction matrix, and determining the direction of the blind element point as the direction of the blind element point according to the consistent direction if the direction of the blind element point is determined to be consistent with any one direction of the four directions; and the replacing module is used for modifying and replacing the pixel values of the blind pixel points by the mean value of the pixel values of the points on the line on which the blind pixel points in the matrix with the determined direction consistent with the direction and the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points in the matrix with the same direction are on line.
The further technical scheme of the invention is as follows: the direction determining module determines the direction of the blind pixel point according to the ratio of the mean value of the pixel values of the points on the line where the blind pixel points are located to the mean value of the pixel values of the points on the line adjacent to the line where the blind pixel points are located in the same direction matrix, wherein the ratio of the mean value of the pixel values of the points on the line is greater than 0.75 and less than 1.25.
The further technical scheme of the invention is as follows: when the replacement module calculates and modifies the pixel values of the replaced blind pixel point, the average ratio of the pixel values of the points on the lines where the blind pixel points in the matrix in the same direction are located is 50%, and the average values of the pixel values of the points on two adjacent lines where the blind pixel points in the matrix in the same direction are located are 25% respectively.
The invention has the technical effects that: a blind pixel image restoration method and system based on a scene are constructed, and a scene image is obtained: acquiring a scene image imaged by a thermal infrared imager; establishing a matrix: establishing a matrix by taking a blind pixel point in a scene image as a center; determining the directionality of the blind spot: four directions are determined with the dummy point as the center, namely: determining the direction of the blind element point according to the ratio of the mean value of the pixel values of the points on the line where the blind element point is located to the mean value of the pixel values of the points on the line adjacent to the line where the blind element point is located in the same direction matrix, and determining the direction of the blind element point as the direction of the blind element point according to the consistent direction if the direction of the blind element point is determined to be consistent with any one direction of the four directions; pixel value replacement: and modifying and replacing the pixel values of the blind pixel points by the mean value of the pixel values of the points on the line on which the blind pixel points in the matrix with the determined direction consistent with the direction and the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points in the matrix with the same direction are on line. The method carries out restoration and replacement through the directions of the blind pixel points and the pixel value mean values in the adjacent directions so as to correct the known blind pixels of the thermal infrared imager and ensure that the known blind pixels have good consistency with the images formed by the thermal infrared imager.
Drawings
FIG. 1 is a schematic diagram of a matrix according to the present invention.
Fig. 2 is a schematic diagram of a module structure according to the present invention.
Detailed Description
The technical solution of the present invention is further illustrated below with reference to specific examples.
The specific implementation mode of the invention is as follows: the method for repairing the blind pixel image based on the scene comprises the following steps:
acquiring a scene image: and acquiring a scene image imaged by the thermal infrared imager.
The specific implementation process is as follows: and imaging through a thermal infrared imager to obtain a scene image formed by the thermal infrared imager.
Establishing a matrix: and establishing a matrix by taking the blind pixel point in the scene image as the center.
The specific implementation process is as follows: and establishing a matrix by taking the blind pixel point in the scene image as the center. In the specific embodiment, a 7 × 7 matrix is taken with the dummy point of the pixel to be replaced as the center point, as shown in fig. 1, D44Is the central dummy point.
Determining the directionality of the blind spot: four directions are determined with the dummy point as the center, namely: and determining the direction of the blind element point according to the ratio of the mean value of the pixel values of the points on the line where the blind element point is located to the mean value of the pixel values of the points on the line adjacent to the line where the blind element point is located in the same direction matrix, and if the direction of the blind element point is determined to be consistent with any one direction of the four directions, determining the consistent direction as the direction of the blind element point.
The preferred embodiments of the present invention are: and when the modification and replacement of the pixel values of the blind pixel point images are determined, performing the modification and replacement of the pixel values of the blind pixel point images by using the mean value of the pixel values of the points on the lines where the blind pixel points in the matrixes in the same direction are located and the mean value of the pixel values of the points on two adjacent lines where the blind pixel points in the matrixes in the same direction are located.
The preferred embodiments of the present invention are: the condition for determining that the direction of the dummy point coincides with any one of the four directions is: the ratio of the mean value of the pixel values of the points on the line on which the blind pixel points are located to the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points are located in the same-direction matrix is greater than 0.75 and less than 1.25. The condition for determining that the direction of the dummy point coincides with any one of the four directions is: the ratio of the mean value of the pixel values of the points on the line on which the blind pixel points are located to the mean value of the pixel values of the points on two lines adjacent to the line on which the blind pixel points are located in the same direction matrix is greater than 0.75 and less than 1.25.
The preferred embodiments of the present invention are: if there are other blind primitive points in the point on the line where the blind primitive point is located, the pixel value of the other blind primitive points is replaced with the pixel value of any non-blind primitive point on the line. If there are other blind primitive points in the points on the line adjacent to the line where the blind primitive point is located, the pixel values of the other blind primitive points are replaced with the pixel values of any non-blind primitive points on the line.
The specific implementation process is as follows:
and verifying the consistency of the first diagonal direction of the matrix and the blind pixel point direction, and determining the blind pixel point direction.
As shown in figure 1 of the drawings, in which,
the direction of a line formed by connecting D11, D22, D33, D44, D55, D66 and D77 is defined as a first diagonal direction of a blind pixel point.
The direction in which D21, D32, D43, D54, D65, and D76 are connected is defined as a first diagonal direction 1.
The direction in which D12, D23, D34, D45, D56, D67 are connected is defined as the first diagonal direction 2.
Calculating the average value of the pixel gray scale of the point in the first diagonal direction of the matrix:
if other blind pixel points exist in D11, D22, D33, D55, D66 and D77, the gray value of the corresponding pixel point can be replaced by the gray value of any non-blind pixel point in the direction.
The average value of the pixel gradations of the points in the first diagonal direction 1 and the first diagonal direction 2 is calculated.
Average value of pixel gray scale of a point in the first diagonal direction 1 of the matrix:
if there are other blind meta-points in D21, D32, D43, D54, D65, D76, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
Average value of pixel gray scale of a point in the first diagonal direction 2 of the matrix:
if there are other blind meta-points in D12, D23, D34, D45, D56, D67, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
If the following conditions are satisfied:
And verifying the consistency of the second diagonal direction of the matrix and the blind pixel point direction, and determining the blind pixel point direction.
The direction in which D17, D26, D35, D44, D53, D62 and D71 are connected is defined as a second diagonal direction.
The direction in which D27, D36, D45, D54, D63, D72 are connected is defined as the second diagonal direction 1.
The direction in which D16, D25, D34, D43, D52, D61 are connected is defined as the second diagonal direction 2.
And calculating the average value of the pixel gray scale of the points in the second diagonal direction, the second diagonal direction 1 and the second diagonal direction 2 of the matrix.
Average pixel gray level of points in the second diagonal direction of the matrix
If there are other blind meta-points in D17, D26, D35, D53, D62, D71, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
Average pixel gray level for a point in the second diagonal direction 1 of the matrix:
if there are other blind meta-points in D27, D36, D45, D54, D63, D72, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
Average pixel intensity for a point in the second diagonal direction 2 of the matrix:
if there are other blind meta-points in D16, D25, D34, D43, D52, D61, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
If the following conditions are satisfied:
And verifying the consistency of the vertical direction of the matrix and the blind pixel point direction, and determining the blind pixel point direction.
The direction of a line formed by connecting D14, D24, D34, D44, D54, D64 and D74 is defined as the vertical direction of the blind pixel.
The direction in which lines D13, D23, D33, D43, D53, D63, and D73 are connected is defined as the vertical direction 1.
The direction in which lines D15, D25, D35, D45, D55, D65, and D75 are connected is defined as the vertical direction 2.
The average value of the pixel gray levels of the dots in the vertical direction of the matrix is calculated.
If there are other blind meta-points in D14, D24, D34, D54, D64, D74, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
Calculate the pixel gray level average for the point in the matrix vertical direction 1:
if other blind meta-points exist in D13, D23, D33, D43, D53, D63, D73, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
Calculate the pixel gray level average for the point in the matrix vertical direction 2:
if other blind meta-points exist in D15, D25, D35, D45, D55, D65, D75, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
If the following conditions are satisfied:
And verifying the consistency of the horizontal direction of the matrix and the blind pixel point direction, and determining the blind pixel point direction.
The direction of a line formed by connecting D41, D42, D43, D44, D45, D46 and D47 is defined as the horizontal direction of the blind spot.
The direction in which lines D31, D32, D33, D34, D35, D36, and D37 are connected is defined as the horizontal direction 1.
The direction in which lines D51, D52, D53, D54, D55, D56, and D57 are connected is defined as the horizontal direction 2.
Calculating the average value of the pixel gray levels of the points in the horizontal direction of the matrix:
if there are other blind meta-points in D41, D42, D43, D45, D46, D47, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
Calculate the pixel gray level average for the point in the horizontal direction 1 of the matrix:
if other blind meta-points exist in D31, D32, D33, D34, D35, D36, D37, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
Calculate the pixel gray level average for the point in the horizontal direction 2 of the matrix:
if other blind meta-points exist in D51, D52, D53, D54, D55, D56, D57, the value of the corresponding point can be replaced by the value of any non-blind meta-point in that direction.
If the following conditions are satisfied:
Pixel value replacement: and modifying and replacing the pixel values of the blind pixel points by the mean value of the pixel values of the points on the line on which the blind pixel points in the matrix with the determined direction consistent with the direction and the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points in the matrix with the same direction are on line.
The specific implementation process is as follows: the method determines the replacement pixel value of the blind pixel point jointly through the mean value of the pixel values of the points on the line where the blind pixel points in the matrix with the same direction of the blind pixel points are located and the mean value of the pixel values of the points on the line adjacent to the line where the blind pixel points in the matrix with the same direction are located. In a specific embodiment, when the modified and replaced pixel values of the blind pixel point are calculated, the average weight ratio of the pixel values of the points on the line where the blind pixel points are located in the matrix in the same direction is greater than the average weight ratio of the pixel values of the points on the line adjacent to the line where the blind pixel points are located in the matrix in the same direction. The average value of the pixel values of the points on the line where the blind pixel points in the matrix with the same direction of the blind pixel points are located and the average value of the pixel values of the points on the line adjacent to the line where the blind pixel points are located in the matrix with the same direction are determined by respectively accounting for a certain percentage, namely: and determining the replacement pixel value of the blind pixel point by adopting fifty percent of the average value of the pixel values of the points on the line on which the blind pixel points in the matrix with the consistent direction of the blind pixel points and fifty percent of the average value of the pixel values of the points on the line adjacent to the blind pixel points in the matrix with the same direction. In a preferred embodiment, when calculating the modified and replaced pixel values of the blind pixel point map, the average ratio of the pixel values of the points on the lines where the blind pixel points are located in the matrix in the same direction is 50%, and the average values of the pixel values of the points on two lines adjacent to the lines where the blind pixel points are located in the matrix in the same direction are 25% each.
The preferred embodiments of the present invention are: when the modified and replaced blind pixel point pixel values are calculated, the average ratio of the pixel values of the points on the lines where the blind pixel points are located in the matrix in the same direction is 50%, and the average values of the pixel values of the points on two adjacent lines where the blind pixel points are located in the matrix in the same direction are respectively 25%.
The specific implementation process is as follows:
if the direction of the blind pixel point to be replaced is consistent with the vertical direction, the replacement result of the blind pixel point is as follows:
D44rep＝0.5*avgvertmid+0.25*avgvert1+0.25*avgvert2
if the direction of the blind pixel point to be replaced is consistent with the horizontal direction, the replacement result of the blind pixel point is as follows:
D44rep＝0.5*avghormid+0.25*avghor1+0.25*avghor2
if the direction of the blind pixel point to be replaced is consistent with the direction of the first diagonal line, the replacing result of the blind pixel point is as follows:
D44rep＝0.5*avgleftmid+0.25*avgleft1+0.25*avgleft2
if the direction of the blind pixel point to be replaced is consistent with the direction of the second diagonal line, the replacing result of the blind pixel point is as follows:
D44rep＝0.5*avgrightmid+0.25*avgright1+0.25*avgright2
as shown in fig. 2, the embodiment of the present invention is: the method comprises the steps of constructing a blind pixel image restoration system based on a scene, wherein the system comprises a scene image acquisition module 1, a matrix establishment module 2, a direction determination module 3 for determining the directionality of blind pixel points and a replacement module 4 for replacing pixel values, wherein the scene image acquisition module 1 acquires a scene image imaged by a thermal infrared imager; the matrix establishing module 2 establishes a matrix by taking a blind pixel point in a scene image as a center; the direction determination module 3 determines four directions with the dummy point as the center, that is: determining the direction of the blind element point according to the ratio of the mean value of the pixel values of the points on the line where the blind element point is located to the mean value of the pixel values of the points on the line adjacent to the line where the blind element point is located in the same direction matrix, and determining the direction of the blind element point as the direction of the blind element point according to the consistent direction if the direction of the blind element point is determined to be consistent with any one direction of the four directions; the replacing module 4 performs modification replacement on the pixel values of the blind pixel points by using the average value of the pixel values of the points on the line where the blind pixel points in the matrix with the determined direction consistent with the direction and the average value of the pixel values of the points on the line adjacent to the line where the blind pixel points in the matrix with the same direction are on the line.
The preferred embodiments of the present invention are: the direction determining module 3 determines the direction of the blind pixel point according to the ratio of the mean value of the pixel values of the points on the line where the blind pixel points are located to the mean value of the pixel values of the points on the line adjacent to the line where the blind pixel points are located in the same direction matrix, which is greater than 0.75 and less than 1.25.
The preferred embodiments of the present invention are: when the replacement module 4 calculates and modifies the pixel values of the replaced blind pixel point, the average ratio of the pixel values of the points on the lines where the blind pixel points in the matrix in the same direction are located is 50%, and the average values of the pixel values of the points on two adjacent lines where the blind pixel points in the matrix in the same direction are located are 25% respectively.
The invention has the technical effects that: a blind pixel image restoration method and system based on a scene are constructed, and a scene image is obtained: acquiring a scene image imaged by a thermal infrared imager; establishing a matrix: establishing a matrix by taking a blind pixel point in a scene image as a center; determining the directionality of the blind spot: four directions are determined with the dummy point as the center, namely: determining the direction of the blind element point according to the ratio of the mean value of the pixel values of the points on the line where the blind element point is located to the mean value of the pixel values of the points on the line adjacent to the line where the blind element point is located in the same direction matrix, and determining the direction of the blind element point as the direction of the blind element point according to the consistent direction if the direction of the blind element point is determined to be consistent with any one direction of the four directions; pixel value replacement: and modifying and replacing the pixel values of the blind pixel points by the mean value of the pixel values of the points on the line on which the blind pixel points in the matrix with the determined direction consistent with the direction and the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points in the matrix with the same direction are on line. The method carries out restoration and replacement through the directions of the blind pixel points and the pixel value mean values in the adjacent directions so as to correct the known blind pixels of the thermal infrared imager and ensure that the known blind pixels have good consistency with the images formed by the thermal infrared imager.
The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, several simple deductions or substitutions can be made without departing from the spirit of the invention, and all shall be considered as belonging to the protection scope of the invention.
Claims (10)
1. A scene-based blind pixel image restoration method is characterized by comprising the following steps:
acquiring a scene image: acquiring a scene image imaged by a thermal infrared imager;
establishing a matrix: establishing a matrix by taking a blind pixel point in a scene image as a center;
determining the directionality of the blind spot: four directions are determined with the dummy point as the center, namely: determining the direction of the blind element point according to the ratio of the mean value of the pixel values of the points on the line where the blind element point is located to the mean value of the pixel values of the points on the line adjacent to the line where the blind element point is located in the same direction matrix, and determining the direction of the blind element point as the direction of the blind element point according to the consistent direction if the direction of the blind element point is determined to be consistent with any one direction of the four directions;
pixel value replacement: and determining a modified replacement value of the pixel value of the blind pixel point jointly by the mean value of the pixel values of the points on the line on which the blind pixel points in the matrix with the consistent direction of the determined blind pixel points are positioned and the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points in the matrix with the same direction are positioned.
2. The scene-based blind pixel image inpainting method according to claim 1, wherein when determining the modification replacement of the blind pixel point image pixel values, the modification replacement values of the blind pixel point image pixel values are determined by using a mean value of the pixel values of the points on the line where the blind pixel points are located in the matrix in the same direction and a mean value of the pixel values of the points on two adjacent lines where the blind pixel points are located in the matrix in the same direction.
3. The method according to claim 1, wherein when the modified and replaced blind pixel values are calculated, the average weight ratio of the pixel values of the points on the line where the blind pixel points are located in the matrix in the same direction is greater than the average weight ratio of the pixel values of the points on the line adjacent to the line where the blind pixel points are located in the matrix in the same direction.
4. The method according to claim 2, wherein when calculating the modified and replaced blind pixel point pixel values, the average weight of the pixel values of the points on the lines where the blind pixel points in the matrix in the same direction are located is 50%, and the average weights of the pixel values of the points on the two lines adjacent to the lines where the blind pixel points in the matrix in the same direction are located are 25%.
5. The scene-based blind pixel image restoration method according to claim 1, wherein the condition for determining that the direction of the blind pixel point is consistent with any one of the four directions is: the ratio of the mean value of the pixel values of the points on the line on which the blind pixel points are located to the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points are located in the same-direction matrix is greater than 0.75 and less than 1.25.
6. The method according to claim 1, wherein if there is another blind primitive point in the points on the line where the blind primitive point is located, the pixel value of the other blind primitive point is replaced with the pixel value of any non-blind primitive point on the line.
7. The method as claimed in claim 1, wherein the matrix is 7X7 matrix centered on the blind pixel point in the scene image.
8. A blind pixel image restoration system based on a scene is characterized by comprising a scene image acquisition module, a matrix establishment module, a direction determination module for determining the directivity of blind pixel points and a replacement module for replacing pixel values, wherein the scene image acquisition module acquires a scene image imaged by a thermal infrared imager; the matrix establishing module establishes a matrix by taking a blind pixel point in a scene image as a center; the direction determination module determines four directions by taking a dummy point as a center, namely: determining the direction of the blind element point according to the ratio of the mean value of the pixel values of the points on the line where the blind element point is located to the mean value of the pixel values of the points on the line adjacent to the line where the blind element point is located in the same direction matrix, and determining the direction of the blind element point as the direction of the blind element point according to the consistent direction if the direction of the blind element point is determined to be consistent with any one direction of the four directions; and the replacing module is used for modifying and replacing the pixel values of the blind pixel points by the mean value of the pixel values of the points on the line on which the blind pixel points in the matrix with the determined direction consistent with the direction and the mean value of the pixel values of the points on the line adjacent to the line on which the blind pixel points in the matrix with the same direction are on line.
9. The scene-based blind pixel image inpainting system according to claim 8, wherein said direction determining module determines the direction of the blind pixel point according to a ratio of a mean value of pixel values of points on a line where the blind pixel point is located to a mean value of pixel values of points on a line adjacent to the blind pixel point in the same direction matrix, which is greater than 0.75 and less than 1.25.
10. The scene-based blind pixel image inpainting system according to claim 8, wherein said replacing module calculates a mean weight ratio of pixel values of points on a line on which the blind pixel points in the matrix in the same direction are located when the pixel values of the replaced blind pixel points are modified, which is greater than a mean weight ratio of pixel values of points on a line adjacent to the line on which the blind pixel points in the matrix in the same direction are located.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811137990.8A CN109360165B (en) | 2018-09-28 | 2018-09-28 | Scene-based blind pixel image restoration method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811137990.8A CN109360165B (en) | 2018-09-28 | 2018-09-28 | Scene-based blind pixel image restoration method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109360165A CN109360165A (en) | 2019-02-19 |
| CN109360165B true CN109360165B (en) | 2020-08-04 |
Family
ID=65348157
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811137990.8A Active CN109360165B (en) | 2018-09-28 | 2018-09-28 | Scene-based blind pixel image restoration method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109360165B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110823385A (en) * | 2019-11-22 | 2020-02-21 | 江苏东方赛光电有限公司 | Efficient algorithm for replacing blind pixels of infrared focal plane |
| CN113038047B (en) * | 2019-12-25 | 2022-09-06 | 中国电子科技集团公司第二十四研究所 | Digital pixel readout circuit, pixel array and image sensor |
| CN111724320B (en) * | 2020-06-19 | 2021-01-08 | 北京波谱华光科技有限公司 | Blind pixel filling method and system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1622637A (en) * | 2004-12-27 | 2005-06-01 | 北京中星微电子有限公司 | Image dead point and noise eliminating method |
| CN101908209A (en) * | 2010-07-29 | 2010-12-08 | 中山大学 | Cubic spline-based infrared thermal image blind pixel compensation algorithm |
| CN103268594A (en) * | 2013-05-17 | 2013-08-28 | 山东神戎电子股份有限公司 | Blind pixel replacing method of thermal infrared imager system |
| CN104463891A (en) * | 2014-12-22 | 2015-03-25 | 西安电子科技大学 | Defective pixel detecting and correcting method based on minimal mid-value gradient filtering |
| CN105069747A (en) * | 2015-07-13 | 2015-11-18 | 中国科学院遥感与数字地球研究所 | Image interpolation method and device |
| CN107248147A (en) * | 2017-06-07 | 2017-10-13 | 上海晔芯电子科技有限公司 | The method that bad point is eliminated in RAW RGB images |
| CN107305695A (en) * | 2016-04-14 | 2017-10-31 | 上海富瀚微电子股份有限公司 | A kind of automatic bad point means for correcting of image and method |
| CN107948634A (en) * | 2017-11-22 | 2018-04-20 | 深圳开阳电子股份有限公司 | A kind of dead pixel points of images detection method, device and picture processing chip |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5533196B2 (en) * | 2010-04-27 | 2014-06-25 | 株式会社リコー | Method for identifying data point distribution region on coordinate plane and identification program thereof |
| US10318128B2 (en) * | 2015-09-30 | 2019-06-11 | Adobe Inc. | Image manipulation based on touch gestures |
| CN105928627B (en) * | 2016-04-08 | 2019-01-08 | 北京航天计量测试技术研究所 | A kind of heteropical method of elimination refrigeration mode thermal imaging system picture reproduction |
-
2018
- 2018-09-28 CN CN201811137990.8A patent/CN109360165B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1622637A (en) * | 2004-12-27 | 2005-06-01 | 北京中星微电子有限公司 | Image dead point and noise eliminating method |
| CN101908209A (en) * | 2010-07-29 | 2010-12-08 | 中山大学 | Cubic spline-based infrared thermal image blind pixel compensation algorithm |
| CN101908209B (en) * | 2010-07-29 | 2012-03-28 | 中山大学 | Cubic spline-based infrared thermal image blind pixel compensation algorithm |
| CN103268594A (en) * | 2013-05-17 | 2013-08-28 | 山东神戎电子股份有限公司 | Blind pixel replacing method of thermal infrared imager system |
| CN103268594B (en) * | 2013-05-17 | 2016-03-23 | 山东神戎电子股份有限公司 | A kind of blind element replacement method of infrared thermal imagery instrument system |
| CN104463891A (en) * | 2014-12-22 | 2015-03-25 | 西安电子科技大学 | Defective pixel detecting and correcting method based on minimal mid-value gradient filtering |
| CN105069747A (en) * | 2015-07-13 | 2015-11-18 | 中国科学院遥感与数字地球研究所 | Image interpolation method and device |
| CN107305695A (en) * | 2016-04-14 | 2017-10-31 | 上海富瀚微电子股份有限公司 | A kind of automatic bad point means for correcting of image and method |
| CN107248147A (en) * | 2017-06-07 | 2017-10-13 | 上海晔芯电子科技有限公司 | The method that bad point is eliminated in RAW RGB images |
| CN107948634A (en) * | 2017-11-22 | 2018-04-20 | 深圳开阳电子股份有限公司 | A kind of dead pixel points of images detection method, device and picture processing chip |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109360165A (en) | 2019-02-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109360165B (en) | Scene-based blind pixel image restoration method and system | |
| JP5701182B2 (en) | Image processing apparatus, image processing method, and computer program | |
| CN1812592A (en) | Method and apparatus for processing image data of a color filter array | |
| EP3340171B1 (en) | Depth image denoising method and denoising device | |
| CN1812595A (en) | Color interpolation algorithm | |
| CN1787615A (en) | System and method for detecting and correcting defective pixels in a digital image sensor | |
| JP6537376B2 (en) | Image processing apparatus and image processing method | |
| CN1885311A (en) | Two-dimensional code, encoding and decoding method thereof | |
| CN101079956A (en) | Image processing apparatus and image processing method | |
| CN1856028A (en) | Camera and method for taking picture compensation | |
| TWI588757B (en) | Edge detection system and methods | |
| CN1893618A (en) | Method and apparatus for processing bayer-pattern color digital image signal | |
| CN1816117A (en) | Image sensor with embedded optical element | |
| JP5701181B2 (en) | Image processing apparatus, image processing method, and computer program | |
| JP2017013380A (en) | Image processing device and image processing method | |
| CN110889827A (en) | Transmission line tower online identification and inclination detection method based on vision | |
| US9092845B2 (en) | Image processing device with defect correcting circuit and image processing method | |
| RU2325044C1 (en) | Gradient method of object contour extraction at halftone raster image matrix | |
| CN105760869B (en) | Examination paper marking method based on image gray level recognition and processing | |
| CN1790378A (en) | Binary method and system for image | |
| CN104537627B (en) | A kind of post-processing approach of depth image | |
| CN101031023A (en) | Method and system for processing image | |
| CN105354813A (en) | Method and device for driving pan-tilt-zoom (PTZ) to generate spliced image | |
| CN113068011B (en) | Image sensor, image processing method and system | |
| CN104980718B (en) | Colour correcting apparatus and method |
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 |