CN115018936A - Image dead pixel removing method, device and correction system based on normal distribution - Google Patents
Image dead pixel removing method, device and correction system based on normal distribution Download PDFInfo
- Publication number
- CN115018936A CN115018936A CN202210602049.9A CN202210602049A CN115018936A CN 115018936 A CN115018936 A CN 115018936A CN 202210602049 A CN202210602049 A CN 202210602049A CN 115018936 A CN115018936 A CN 115018936A
- Authority
- CN
- China
- Prior art keywords
- pixel
- point
- detected
- dead
- channel image
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000012216 screening Methods 0.000 claims abstract description 41
- 238000000926 separation method Methods 0.000 claims abstract description 14
- 102100037651 AP-2 complex subunit sigma Human genes 0.000 claims description 6
- 101000806914 Homo sapiens AP-2 complex subunit sigma Proteins 0.000 claims description 6
- 230000002093 peripheral effect Effects 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims 1
- 230000000694 effects Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 230000014509 gene expression Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- 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
- G06T2207/10024—Color image
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30121—CRT, LCD or plasma display
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a device and a correction system for removing image dead pixels based on normal distribution, wherein the method comprises the following steps: performing RGB channel separation on an image to be processed to obtain an R, G, B channel image; performing dead pixel removal processing on each channel image; the dead pixel removal processing includes: calculating the pixel average value mu and the pixel standard deviation sigma of the channel image by using the pixel value of each pixel point in the channel image after the background pixel points are removed; screening out pixel points suspected to be dead points as candidate points to be detected according to the mu and the sigma; judging whether the candidate point to be detected is a bad point or not according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected; and if the candidate point to be detected is a dead point, correcting the pixel values of the pixel points in the preset neighborhood of the candidate point to be detected and the candidate point to be detected. The method, the device and the correction system provided by the invention can effectively improve the dead pixel removal efficiency and the dead pixel identification accuracy.
Description
Technical Field
The invention relates to the field of LED display screen correction, in particular to a method, a device and a correction system for removing image dead pixels based on normal distribution.
Background
In the process of correcting the LED display screen, an upper computer is required to shoot through a camera to obtain a lamp point image as a correction image, and a correction coefficient is calculated through the correction image.
However, due to the use loss and the like, a problem may occur in a particular light sensing device in the camera imaging module, which causes a pixel with an abnormal color in the collected light point image (that is, a dead pixel exists in the light point image collected by the camera due to the damage of the light sensing device in the camera), and the dead pixel may sometimes appear in the center of the light point image, for example, the light point image shown in fig. 1 is the light point image with the dead pixel (wherein the P, Q indicating position is the dead pixel position), and if the collected light point image is abnormal, the correction coefficient calculated by using the light point image subsequently may cause the calculated correction coefficient to be inaccurate, thereby seriously affecting the correction effect of the LED display screen.
Some methods for removing dead pixels in the prior art have more problems, and cannot well remove dead pixels in a corrected image, for example, the dead pixel removal efficiency is low due to the fact that the required calculation amount is too high when the dead pixels are removed; in addition, the accuracy of identifying the dead pixel is not high, which affects the correction effect of the LED display screen and the display effect of the LED display screen.
Disclosure of Invention
In order to solve the above problems, the present invention provides a method, an apparatus, and a system for removing image dead pixels based on normal distribution. The method comprises the following steps:
performing RGB channel separation on an image to be processed to obtain an R, G, B channel image;
performing dead pixel removal processing on each channel image; the dead pixel removal processing includes:
calculating the pixel average value mu and the pixel standard deviation sigma of the channel image by using the pixel value of each pixel point in the channel image after the background pixel point is removed;
screening out pixel points suspected to be dead points as candidate points to be detected according to the mu and the sigma;
judging whether the candidate point to be detected is a bad point or not according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected;
and if the candidate point to be detected is a dead point, correcting the pixel values of the pixel points in the preset neighborhood of the candidate point to be detected and the candidate point to be detected so as to remove the dead point.
Further, the dead pixel removing process further comprises identifying background pixel points in the channel image by a trigonometry, wherein identifying the background pixel points in the channel image by the trigonometry comprises:
counting the pixel distribution condition of the channel image to obtain a pixel distribution condition image, wherein the abscissa of the pixel distribution condition image is a pixel value, and the ordinate is the number of pixel points;
determining the highest point and the lowest point in the pixel distribution situation graph, and connecting the highest point and the lowest point to form a straight line;
taking a pixel value corresponding to a point which is farthest from the straight line in the pixel distribution situation graph as a pixel threshold value;
and determining the pixel points with the pixel values smaller than the pixel threshold value as background pixel points.
Further, screening out the pixel points suspected as dead points as candidate points to be detected according to the mu and the sigma comprises:
according to a preset pixel point selection sequence, sequentially selecting a pixel point from the channel image as a target pixel point, and performing screening operation on the target pixel point until the screening operation is completed on all pixel points in the channel image so as to screen out the pixel point suspected as a dead point as a candidate point to be detected;
wherein the screening operation comprises: for the target pixel point, judging whether the pixel value of the target pixel point in the channel image is greater than mu + sigma;
judging whether the target pixel point is the highest point of the local pixel value in the preset neighborhood range in the channel image;
and if the pixel value of the target pixel point in the channel image is greater than mu + sigma and the target pixel point is the highest point of the local pixel value in the preset neighborhood range in the channel image, taking the target pixel point as a candidate point to be detected.
Further, judging whether the candidate point to be detected is a dead point according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected comprises:
calculating the pixel average value mu 1 and the pixel standard deviation sigma 1 of pixel points in a preset neighborhood range of the candidate point to be detected in the channel image;
acquiring the color of an image to be processed and the color of a channel image, and judging whether the channel image is a main channel image or not according to the color of the image to be processed and the color of the channel image, wherein the color of the R channel image is red, the color of the G channel image is green, and the color of the B channel image is blue;
and if the channel image is the main channel image, judging whether the pixel value of the candidate point to be detected in the channel image is greater than mu 1+ (3 x sigma 1), and if so, judging that the candidate point to be detected is a bad point.
Further, judging whether the candidate point to be detected is a dead point according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected further comprises:
if the channel image is an auxiliary channel image, acquiring a pixel value mu 2 of a candidate point to be detected in the main channel image
Calculating a pixel standard deviation sigma 2 of a pixel point in a preset neighborhood range of a candidate point to be detected in the main channel image;
and judging whether the pixel value of the candidate point to be detected in the channel image is greater than mu 1+ (3 x sigma 1) and greater than mu 2-sigma 2, if so, judging that the candidate point to be detected is a dead point.
Further, the preset neighborhood range is an 8-neighborhood with the candidate point to be detected as the center.
Further, the step of correcting the pixel values of the pixel points in the preset neighborhood of the candidate point to be detected and the candidate point to be detected comprises the following steps:
acquiring pixel values of correction pixel points corresponding to candidate points to be detected in a channel image, wherein the correction pixel points are peripheral range pixel points of pixel points in a preset neighborhood range of the candidate points to be detected;
calculating the pixel average value mu 3 of the correction pixel points in the channel image;
and adjusting the pixel value of the candidate point to be detected in the channel image and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected in the channel image to be mu 3.
The invention also provides an image dead pixel removing device based on normal distribution, which comprises an image separating module and a dead pixel processing module, wherein:
the image separation module is connected with the dead pixel processing module and is used for performing RGB channel separation on the image to be processed to obtain R, G, B channel images;
the dead pixel processing module is used for removing dead pixels aiming at each channel image and comprises a candidate point screening unit to be detected, a dead pixel judging unit and a dead pixel correcting unit; the candidate point screening unit to be detected is connected with the dead pixel judgment unit and is used for calculating the pixel average value mu and the pixel standard deviation sigma of the channel image by using the pixel value of each pixel point in the channel image after the background pixel point is removed; screening out pixel points suspected to be dead points as candidate points to be detected according to the mu and the sigma;
the dead pixel judging unit is connected with the dead pixel correcting unit and used for judging whether the candidate point to be detected is a dead pixel or not according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected;
and the dead pixel correction unit is used for correcting the pixel values of the pixel points in the preset neighborhood of the candidate point to be detected and the candidate point to be detected when the candidate point to be detected is judged to be the dead pixel so as to remove the dead pixel.
Further, the candidate point screening unit for detecting screens out the pixel points suspected to be dead pixels as candidate points to be detected according to the μ and the σ, and the candidate point screening unit for detecting the dead pixels comprises:
according to a preset pixel point selection sequence, sequentially selecting a pixel point from the channel image as a target pixel point, and performing screening operation on the target pixel point until the screening operation is completed on all pixel points in the channel image so as to screen out the pixel point suspected as a dead point as a candidate point to be detected;
wherein, the screening operation includes:
for the target pixel point, judging whether the pixel value of the target pixel point in the channel image is greater than mu + sigma;
judging whether the target pixel point is the highest point of the local pixel value in the preset neighborhood range in the channel image;
and if the pixel value of the target pixel point in the channel image is greater than mu + sigma and the target pixel point is the highest point of the local pixel value in the preset neighborhood range in the channel image, taking the target pixel point as a candidate point to be detected.
The utility model provides a correction system, includes camera, host computer, card and LED display screen send, wherein:
the camera is connected with the upper computer and used for shooting a correction image of the LED display screen and sending the correction image to the upper computer;
the upper computer is connected with the sending card and used for removing dead pixels in the corrected image according to the image dead pixel removing method based on normal distribution, calculating a correction coefficient of each lamp point in the LED display screen according to the corrected image after dead pixels are removed, and sending the correction coefficient to the sending card;
the sending card is connected with the receiving card of the LED display screen and used for sending the correction coefficient to the receiving card of the LED display screen;
and the LED display screen is used for correcting the display data according to the received correction coefficient of each lamp point through the receiving card and displaying the corrected display data.
The image dead pixel removing method, the image dead pixel removing device and the image dead pixel correcting system based on normal distribution, which are provided by the invention, at least have the following beneficial effects:
(1) before dead pixel identification, suspected dead pixels are screened out by using the pixel average value mu and the pixel standard deviation sigma of the channel image, and then the suspected dead pixels are judged one by one to identify whether the suspected dead pixels are dead pixels, so that dead pixel identification of all pixel points in the channel image is not needed, the calculated amount is greatly reduced, and the dead pixel identification efficiency of the channel image is improved;
(2) before calculating the pixel average value mu and the pixel standard deviation sigma of the channel image, identifying background pixel points in the channel image, calculating the pixel average value mu and the pixel standard deviation sigma of the channel image by using the pixel values of all the pixel points in the channel image after the background pixel points are removed, reducing or even eliminating the influence of the background pixel points on the pixel average value and the pixel standard deviation of the channel image, improving the accuracy of suspected dead pixel identification, and further improving the accuracy of dead pixel identification;
(3) after the suspected dead pixel is determined to be used as the candidate point to be detected, judging whether the candidate point to be detected is a dead pixel or not according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected, namely judging that data used by the dead pixel are specific and suitable for the candidate point to be detected aiming at each candidate point to be detected, so that the dead pixel identification accuracy can be further improved;
(4) after identifying the dead pixel, because the dead pixel also has an influence on the pixel value of surrounding pixel points, not only the pixel value of the dead pixel is corrected in this embodiment, but also the pixel value of the pixel point in the preset neighborhood of the dead pixel is corrected, thereby eliminating the influence of the dead pixel on the corrected image to the maximum extent, being beneficial to improving the correction effect of the display screen, and further improving the display effect of the display screen.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic view of a lamp spot image with a dead spot;
FIG. 2 is a flowchart illustrating the steps of a method for removing dead pixels from an image based on normal distribution according to an embodiment of the present invention;
FIG. 3 is a flowchart of the method of step S102 in FIG. 2;
FIG. 4 is a schematic diagram of a channel image pixel distribution;
FIG. 5 is a flowchart of the screening operation of step S203 in FIG. 3;
FIG. 6 is a schematic diagram of a candidate point to be detected and pixel points in a neighborhood range of 8 candidate points to be detected;
FIG. 7 is a flowchart of the method of step S204 in FIG. 3;
FIG. 8 is a flowchart of the method of step S205 in FIG. 3;
FIG. 9 is a schematic diagram of peripheral pixels of pixels in an 8-neighborhood range centered on a candidate point to be detected;
FIG. 10 is a schematic structural diagram of an image dead pixel removing apparatus based on normal distribution according to an embodiment of the present invention;
FIG. 11 is a schematic diagram of a calibration system according to an embodiment of the present invention;
the method comprises the steps of 1-image separation module, 2-dead pixel processing module, 201-background pixel point identification unit, 202-candidate point screening unit to be detected, 203-dead pixel judgment unit, 204-dead pixel correction unit, 3-camera, 4-upper computer, 5-sending card and 6-LED display screen.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is to be understood that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In an embodiment of the present invention, as shown in fig. 2, an image dead pixel removing method based on normal distribution is disclosed, wherein the image dead pixel in the present invention refers to a dead pixel in a shot light image caused by a problem of a single photosensitive device in a camera imaging module.
The method comprises the following steps:
step S101: and performing RGB channel separation on the image to be processed to obtain an R, G, B channel image.
The to-be-processed image is a lamp point image obtained by shooting with a camera after the LED display screen is turned on, and preferably, in this embodiment, the camera is a canon camera, that is, the to-be-processed image is obtained by shooting with a canon camera.
When the LED display screen is corrected, the screen color of the LED display screen is adjusted to be red, green and blue, corresponding lamp point images are obtained through camera shooting and collecting and serve as images to be processed, and the screen color of the LED display screen can be controlled to be lightened through upper computer software.
The to-be-processed image obtained by the camera exists R, G, B three channels. When the screen color is red, the color of the acquired image to be processed is red, the main channel is an R channel, and the auxiliary channel is an G, B channel; similarly, when the color of the screen is green, the color of the acquired image to be processed is green, the main channel is a G channel, and the auxiliary channel is an R, B channel; when the screen color is blue, the color of the acquired image to be processed is blue, the main channel is a B channel, and the auxiliary channel is an R, G channel.
In this step, the RGB channel separation is performed on the image to be processed, so that an R, G, B channel image can be obtained, and then the dead pixel removal processing is performed on each channel image.
Step S102: and performing dead pixel removal processing on each channel image.
Further, as shown in fig. 3, for one channel image, the dead pixel removal processing includes the following steps:
step S201: and identifying background pixel points in the channel image.
Wherein, the background pixel points in the channel image can be identified by Otsu method or trigonometry. Specifically, the pixel distribution condition of the channel image is counted to obtain a pixel distribution condition graph (as shown in fig. 4, the abscissa represents the pixel value, and the ordinate represents the number of the pixel points), a pixel threshold is determined by the Otsu method or the trigonometry, and the pixel point with the pixel value smaller than the pixel threshold is determined as the background pixel point.
Preferably, in this step, a triangle method is used to identify background pixel points in the channel image, and identifying the background pixel points in the channel image by the triangle method includes:
counting the pixel distribution condition of the channel image to obtain a pixel distribution condition graph, as shown in fig. 4, the abscissa of the pixel distribution condition graph is the pixel value, and the ordinate is the number of pixels;
determining the highest point and the lowest point in the pixel distribution situation graph, and connecting the highest point and the lowest point to form a straight line;
taking a pixel value corresponding to a point which is farthest from the straight line in the pixel distribution situation graph as a pixel threshold value;
and determining the pixel points with the pixel values smaller than the pixel threshold value as background pixel points.
Step S202: calculating the pixel average value mu and the pixel standard deviation sigma of the channel image by using the pixel value of each pixel point in the channel image after the background pixel points are removed;
step S203: and screening out the suspected pixel points as the candidate points to be detected according to the mu and the sigma.
In this step, the average value μ and standard deviation σ of the pixels of the channel image are used to screen out the pixels suspected to be dead pixels from all the pixels in the channel image as the pixels to be detected.
Step S204: judging whether the candidate point to be detected is a bad point or not according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected;
step S205: and if the candidate point to be detected is a dead point, correcting the pixel values of the pixel points in the preset neighborhood of the candidate point to be detected and the candidate point to be detected so as to remove the dead point.
It should be understood that steps S201 to S205 are processing performed on one channel image, and when steps S201 to S205 are performed on R, G, B for each channel image, the dead pixel removal for the image to be processed can be completed.
According to the image dead pixel removing method based on normal distribution, before dead pixel identification is carried out, suspected dead pixels are screened out by using the pixel average value mu and the pixel standard deviation sigma of the channel image, then the suspected dead pixels are judged one by one, and whether the suspected dead pixels are dead pixels is identified, so that dead pixel identification is not required to be carried out on all pixel points in the channel image, the calculated amount is greatly reduced, and the efficiency of dead pixel identification of the channel image is improved; meanwhile, before calculating the pixel average value mu and the pixel standard deviation sigma of the channel image, identifying background pixel points in the channel image, calculating the pixel average value mu and the pixel standard deviation sigma of the channel image by using the pixel values of all the pixel points in the channel image after the background pixel points are removed, reducing or even eliminating the influence of the background pixel points on the pixel average value and the pixel standard deviation of the channel image, improving the accuracy of suspected dead pixel identification, and further improving the accuracy of dead pixel identification; in addition, after the suspected dead pixel is determined to be used as the candidate point to be detected, whether the candidate point to be detected is a dead pixel or not is judged according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected, namely, for each candidate point to be detected, the data used for judging the dead pixel are all specific and suitable for the candidate point to be detected, so that the dead pixel identification accuracy can be further improved; finally, after the dead pixel is identified, the dead pixel also has influence on the pixel values of the surrounding pixel points, so that the pixel values of the dead pixel are corrected, and the pixel values of the pixel points in the preset neighborhood of the dead pixel are corrected, so that the influence of the dead pixel on the corrected image is eliminated to the maximum extent, the correction effect of the display screen is improved, and the display effect of the display screen is improved.
In another embodiment of the present invention, step S203 is: and according to a preset pixel point selection sequence, sequentially selecting a pixel point from the channel image as a target pixel point, and performing screening operation on the target pixel point until the screening operation is completed on all pixel points in the channel image so as to screen out the pixel points suspected as dead points as candidate points to be detected.
The pixel selection sequence is preset by a technician, and the invention is not limited to this.
As shown in fig. 5, for the target pixel, the screening operation includes the following steps:
step S2031: judging whether the pixel value of the target pixel point in the channel image is greater than mu + sigma;
step S2032: judging whether the target pixel point is the highest point of the local pixel value in the preset neighborhood range in the channel image;
as shown in fig. 6, the preset neighborhood range may be an 8-neighborhood range centered on the target pixel P, that is, the pixels in the preset neighborhood range of the target pixel P are 8 pixels at the position marked as a in fig. 6.
The pixel value of the target pixel point is the highest point of the local pixel value, namely the pixel value of the target pixel point is larger than the pixel values of 8 pixel points in an 8-neighborhood range taking the target pixel point as the center.
Because the dead pixel is usually in the middle of the lamp point image, the situation at the edge is basically very few, if a certain pixel point is located at the most edge position of the lamp point image and there is no 8-neighborhood pixel point taking the pixel point as the center, the pixel point can be directly considered as not being a suspected dead pixel.
Step S2033: and if the pixel value of the target pixel point in the channel image is greater than mu + sigma and the target pixel point is the highest point of the local pixel value in the preset neighborhood range in the channel image, taking the target pixel point as a candidate point to be detected.
In still another embodiment of the present invention, as shown in fig. 7, the step S204 includes the steps of:
step S2041: and calculating the pixel average value mu 1 and the pixel standard deviation sigma 1 of the pixel points in the preset neighborhood range of the candidate points to be detected in the channel image.
Similarly, if the preset neighborhood range is an 8-neighborhood region centered on the candidate point to be detected, that is, the pixel values of 8 pixel points corresponding to the 8-neighborhood region in the channel image are used to calculate the pixel average value μ 1 and the pixel standard deviation σ 1.
Step S2042: and judging whether the channel image is a main channel image, if so, executing step S2043, otherwise, executing step S2044-step S2045.
Specifically, in this step, the color of the image to be processed and the color of the channel image need to be obtained, and whether the channel image is the main channel image is determined according to the color of the image to be processed and the color of the channel image. Wherein, the R channel image color is red, the G channel image color is green, and the B channel image color is blue.
And if the color of the image to be processed is red, the R channel image corresponding to the image to be processed is the main channel image, and the other same principles are carried out. Therefore, whether the channel image being processed is the main channel image can be determined by the color of the image to be processed.
Step S2043: and judging whether the pixel value of the candidate point to be detected in the channel image is greater than mu 1+ (3 x sigma 1), if so, judging that the candidate point to be detected is a bad point.
Step S2044: and acquiring a pixel value mu 2 of the candidate point to be detected in the main channel image, and calculating a pixel standard deviation sigma 2 of a pixel point in the preset neighborhood range of the candidate point to be detected in the main channel image.
Similarly, if the preset neighborhood range is an 8-neighborhood region centered on the candidate point to be detected, that is, the pixel value of 8 pixel points corresponding to the 8-neighborhood region in the main channel image is used to calculate the pixel standard deviation σ 2.
Specifically, in one implementation of the present invention, the R, G, B channel images are simultaneously processed to remove dead pixels, so the main channel image used in this step is the original main channel image without dead pixel value correction.
Step S2045: and judging whether the pixel value of the candidate point to be detected in the channel image is greater than mu 1+ (3 x sigma 1) and greater than mu 2-sigma 2, if so, judging that the candidate point to be detected is a dead point.
When the currently processed image is the auxiliary channel image, the above step S2044 is performed, and it is determined whether the pixel value of the candidate point to be detected in the channel image is greater than μ 2- σ 2, so as to determine whether the auxiliary channel value is close to the main channel value.
In still another embodiment of the present invention, as shown in fig. 8, step S205 includes the steps of:
step S2051: and acquiring the pixel value of a correction pixel point corresponding to the candidate point to be detected in the channel image, wherein the correction pixel point is a peripheral range pixel point of a pixel point in a preset neighborhood range of the candidate point to be detected.
As shown in fig. 9, the preset neighborhood range is: taking the 8-neighborhood range with the candidate point P to be detected as the center as an example, the peripheral range pixel point of the pixel point a in the 8-neighborhood range refers to 16 pixel points that are adjacent to the pixel point in the 8-neighborhood range and are located in the peripheral range of the pixel point in the 8-neighborhood range, that is, 16 pixel points marked as B in fig. 9.
Step S2052: calculating the pixel average value mu 3 of the correction pixel points in the channel image;
step S2053: and adjusting the pixel value of the candidate point to be detected in the channel image and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected in the channel image to be mu 3.
As shown in fig. 10, the present invention further provides an image dead pixel removing apparatus based on normal distribution, the apparatus includes an image separation module 1 and a dead pixel processing module 2, wherein:
the image separation module 1 is connected with the dead pixel processing module 2 and is used for performing RGB channel separation on the image to be processed to obtain R, G, B channel images;
and the dead pixel processing module 2 is used for performing dead pixel removing processing on each channel image.
The dead pixel processing module 2 comprises a candidate point screening unit 202 to be detected, a dead pixel judging unit 203 and a dead pixel correcting unit 204;
the candidate point screening unit 202 to be detected is connected to the dead pixel judgment unit 203, and is configured to calculate a pixel average value μ and a pixel standard deviation σ of the channel image by using the pixel value of each pixel point in the channel image from which the background pixel point is removed; screening out pixel points suspected to be dead points as candidate points to be detected according to the mu and the sigma;
a dead pixel judgment unit 203, connected to the dead pixel correction unit 204, configured to judge whether the candidate point to be detected is a dead pixel according to the pixel value of the candidate point to be detected and the pixel value of a pixel point within a preset neighborhood range of the candidate point to be detected;
the dead pixel correction unit 204 is configured to, when it is determined that the candidate point to be detected is a dead pixel, correct pixel values of pixel points in a preset neighborhood of the candidate point to be detected and the candidate point to be detected, so as to remove the dead pixel.
In another embodiment of the present invention, the bad point processing module 2 further includes a background pixel point identifying unit 201, which is connected to the candidate point screening unit 202 to be detected, and is configured to identify the background pixel points in the channel image, specifically, to identify the background pixel points in the channel image through the tsu method or the trigonometry method.
In another embodiment of the present invention, the step of screening the pixel points suspected to be the bad points by the candidate point screening unit 202 according to μ and σ includes:
according to a preset pixel point selection sequence, sequentially selecting a pixel point from the channel image as a target pixel point, and performing screening operation on the target pixel point until the screening operation is completed on all pixel points in the channel image so as to screen out the pixel point suspected as a dead point as a candidate point to be detected;
wherein the screening operation comprises:
for the target pixel point, judging whether the pixel value of the target pixel point in the channel image is greater than mu + sigma;
judging whether the target pixel point is the highest point of the local pixel value in the preset neighborhood range in the channel image;
and if the pixel value of the target pixel point in the channel image is greater than mu + sigma and the target pixel point is the highest point of the local pixel value in the preset neighborhood range in the channel image, taking the target pixel point as a candidate point to be detected.
In another embodiment of the present invention, the determining, by the dead pixel determining unit 203, whether the candidate point to be detected is a dead pixel according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected includes:
calculating the pixel average value mu 1 and the pixel standard deviation sigma 1 of pixel points in a preset neighborhood range of the candidate point to be detected in the channel image;
acquiring the color of an image to be processed and the color of a channel image, and judging whether the channel image is a main channel image or not according to the color of the image to be processed and the color of the channel image, wherein the color of the R channel image is red, the color of the G channel image is green, and the color of the B channel image is blue;
and if the channel image is the main channel image, judging whether the pixel value of the candidate point to be detected in the channel image is greater than mu 1+ (3 x sigma 1), and if so, judging that the candidate point to be detected is a dead point.
Further, if the channel image is an auxiliary channel image, a pixel value mu 2 of the candidate point to be detected in the main channel image is obtained, a pixel standard deviation sigma 2 of a pixel point in the preset neighborhood range of the candidate point to be detected in the main channel image is calculated, whether the pixel value of the candidate point to be detected in the channel image is larger than mu 1+ (3 x sigma 1) and larger than mu 2-sigma 2 or not is judged, and if yes, the candidate point to be detected is judged to be a bad point.
In an embodiment of the present invention, as shown in fig. 11, there is further provided a calibration system, which includes a camera 3, an upper computer 4, a transmitting card 5 and an LED display screen 6, wherein:
the camera 3 is connected with the upper computer 4 and used for shooting a correction image of the LED display screen 6 and sending the correction image to the upper computer 4;
the upper computer 4 is connected with the sending card 5 and is used for removing dead pixels in the corrected image according to the normal distribution-based image dead pixel removing method, calculating a correction coefficient of each lamp point in the LED display screen 6 according to the corrected image after dead pixels are removed, and sending the correction coefficient to the sending card 5;
the sending card 5 is connected with a receiving card of the LED display screen 6 and is used for sending the correction coefficient to the receiving card of the LED display screen 6;
and the LED display screen 6 is used for correcting the display data according to the received correction coefficient of each lamp point through the receiving card and displaying the corrected display data.
According to the image dead pixel removing method, device and correction system based on normal distribution, before dead pixel identification, suspected dead pixels are screened out by using the pixel average value mu and the pixel standard deviation sigma of the channel image, and then the suspected dead pixels are judged one by one to identify whether the suspected dead pixels are dead pixels, so that dead pixel identification of all pixel points in the channel image is not needed, the calculated amount is greatly reduced, and the efficiency of dead pixel identification of the channel image is improved; meanwhile, before calculating the pixel average value mu and the pixel standard deviation sigma of the channel image, identifying background pixel points in the channel image, calculating the pixel average value mu and the pixel standard deviation sigma of the channel image by using the pixel values of all the pixel points in the channel image after the background pixel points are removed, reducing or even eliminating the influence of the background pixel points on the pixel average value and the pixel standard deviation of the channel image, improving the accuracy of suspected dead pixel identification, and further improving the accuracy of dead pixel identification; in addition, after the suspected dead pixel is determined to be used as the candidate point to be detected, whether the candidate point to be detected is a dead pixel or not is judged according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected, namely, for each candidate point to be detected, the data used for judging the dead pixel are all specific and suitable for the candidate point to be detected, so that the dead pixel identification accuracy can be further improved; finally, after the dead pixel is identified, the dead pixel also has influence on the pixel values of the surrounding pixel points, so that the pixel values of the dead pixel are corrected, and the pixel values of the pixel points in the preset neighborhood of the dead pixel are corrected, so that the influence of the dead pixel on the corrected image is eliminated to the maximum extent, the correction effect of the display screen is improved, and the display effect of the display screen is improved.
The terms and expressions used in the specification of the present invention have been set forth for illustrative purposes only and are not meant to be limiting. It will be appreciated by those skilled in the art that changes could be made to the details of the above-described embodiments without departing from the underlying principles thereof. The scope of the invention is, therefore, indicated by the appended claims, in which all terms are intended to be interpreted in their broadest reasonable sense unless otherwise indicated.
Claims (10)
1. An image dead pixel removing method based on normal distribution is characterized by comprising the following steps:
performing RGB channel separation on an image to be processed to obtain an R, G, B channel image;
performing dead pixel removal processing on each channel image; the dead pixel removal processing includes:
calculating the pixel average value mu and the pixel standard deviation sigma of the channel image by using the pixel value of each pixel point in the channel image after the background pixel point is removed;
screening out pixel points suspected to be dead points as candidate points to be detected according to the mu and the sigma;
judging whether the candidate point to be detected is a bad point or not according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected;
and if the candidate point to be detected is a dead point, correcting the pixel values of the candidate point to be detected and the pixel points in the preset neighborhood of the candidate point to be detected so as to remove the dead point.
2. The method for removing the dead pixel of the image based on the normal distribution as claimed in claim 1, wherein the dead pixel removing process further comprises identifying a background pixel point in the channel image by a trigonometry method;
wherein the identifying background pixel points in the channel image by triangulation comprises:
counting the pixel distribution condition of the channel image to obtain a pixel distribution condition graph, wherein the abscissa of the pixel distribution condition graph is a pixel value, and the ordinate is the number of pixel points;
determining the highest point and the lowest point in the pixel distribution situation graph, and connecting the highest point and the lowest point to form a straight line;
taking a pixel value corresponding to a point which is farthest from the straight line in the pixel distribution situation graph as a pixel threshold value;
and determining the pixel points with the pixel values smaller than the pixel threshold value as background pixel points.
3. The method for removing dead pixels of an image based on normal distribution according to claim 1, wherein the step of screening out pixel points suspected to be dead pixels as candidate points to be detected according to μ and σ comprises:
according to a preset pixel point selection sequence, sequentially selecting a pixel point from the channel image as a target pixel point, and performing screening operation on the target pixel point until the screening operation is completed on all pixel points in the channel image so as to screen out the pixel point suspected as a dead point as a candidate point to be detected;
wherein the screening operation comprises:
for the target pixel point, judging whether the pixel value of the target pixel point in the channel image is greater than mu + sigma;
judging whether the target pixel point is the highest point of the local pixel value in a preset neighborhood range in the channel image;
and if the pixel value of the target pixel point in the channel image is greater than mu + sigma and the target pixel point is the highest point of the local pixel value in the preset neighborhood range in the channel image, taking the target pixel point as a candidate point to be detected.
4. The method for removing the dead pixel from the image based on the normal distribution as claimed in claim 1, wherein the determining whether the candidate point to be detected is a dead pixel according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected comprises:
calculating the pixel average value mu 1 and the pixel standard deviation sigma 1 of the pixel points in the channel image in the preset neighborhood range of the candidate point to be detected;
acquiring the color of the image to be processed and the color of the channel image, and judging whether the channel image is a main channel image or not according to the color of the image to be processed and the color of the channel image, wherein the color of the R channel image is red, the color of the G channel image is green, and the color of the B channel image is blue;
and if the channel image is a main channel image, judging whether the pixel value of the candidate point to be detected in the channel image is greater than mu 1+ (3 x sigma 1), and if so, judging that the candidate point to be detected is a dead point.
5. The method for removing image dead pixels based on normal distribution according to claim 4, wherein the determining whether the candidate point to be detected is a dead pixel according to the pixel value of the candidate point to be detected and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected further comprises:
if the channel image is an auxiliary channel image, acquiring a pixel value mu 2 of the candidate point to be detected in the main channel image;
calculating the pixel standard deviation sigma 2 of pixel points in the main channel image in the preset neighborhood range of the candidate points to be detected;
and judging whether the pixel value of the candidate point to be detected in the channel image is greater than mu 1+ (3 x sigma 1) and greater than mu 2-sigma 2, if so, judging that the candidate point to be detected is a bad point.
6. The method for removing the dead pixel from the image based on the normal distribution as claimed in claim 1, wherein the predetermined neighborhood range is 8 neighborhoods centered on the candidate point to be detected.
7. The method for removing the dead pixel of the image based on the normal distribution as claimed in claim 1, wherein the correcting the candidate point to be detected and the pixel values of the pixel points in the preset neighborhood of the candidate point to be detected comprises:
acquiring pixel values of correction pixel points corresponding to the candidate points to be detected in the channel image, wherein the correction pixel points are peripheral range pixel points of pixel points in a preset neighborhood range of the candidate points to be detected;
calculating the pixel average value mu 3 of the correction pixel points in the channel image;
and adjusting the pixel value of the candidate point to be detected in the channel image and the pixel value of the pixel point in the preset neighborhood range of the candidate point to be detected in the channel image to be mu 3.
8. An image dead pixel removing device based on normal distribution is characterized by comprising an image separation module and a dead pixel processing module, wherein:
the image separation module is connected with the dead pixel processing module and is used for performing RGB channel separation on the image to be processed to obtain R, G, B channel images;
the dead pixel processing module is used for removing dead pixels of each channel image and comprises a candidate point screening unit to be detected, a dead pixel judging unit and a dead pixel correcting unit;
the candidate point screening unit to be detected is connected with the dead pixel judgment unit and is used for calculating the pixel average value mu and the pixel standard deviation sigma of the channel image by using the pixel value of each pixel point in the channel image after the background pixel point is removed; screening out pixel points suspected to be dead points as candidate points to be detected according to the mu and the sigma;
the dead pixel judging unit is connected with the dead pixel correcting unit and is used for judging whether the candidate point to be detected is a dead pixel or not according to the pixel value of the candidate point to be detected and the pixel values of the pixel points in the preset neighborhood range of the candidate point to be detected;
and the dead pixel correction unit is used for correcting the pixel values of the candidate point to be detected and the pixel points in the preset neighborhood of the candidate point to be detected when the candidate point to be detected is judged to be a dead pixel, so as to remove the dead pixel.
9. The normally distributed image dead pixel removing device according to claim 8, wherein the candidate point to be detected screening unit screens out pixel points suspected to be dead pixels as candidate points to be detected according to μ and σ, and includes:
according to a preset pixel point selection sequence, sequentially selecting a pixel point from the channel image as a target pixel point, and performing screening operation on the target pixel point until the screening operation is completed on all pixel points in the channel image so as to screen out the pixel point suspected as a dead point as a candidate point to be detected;
wherein the screening operation comprises: for the target pixel point, judging whether the pixel value of the target pixel point in the channel image is greater than mu + sigma;
judging whether the target pixel point is the highest point of the local pixel value in a preset neighborhood range in the channel image;
and if the pixel value of the target pixel point in the channel image is greater than mu + sigma and the target pixel point is the highest point of the local pixel value in the preset neighborhood range in the channel image, taking the target pixel point as a candidate point to be detected.
10. The utility model provides a correction system, its characterized in that, the system includes camera, host computer, transmission card and LED display screen, wherein:
the camera is connected with the upper computer and used for shooting a corrected image of the LED display screen and sending the corrected image to the upper computer;
the upper computer is connected with the sending card and is used for removing dead pixels in the corrected image according to the normal distribution-based image dead pixel removing method of any one of claims 1 to 7, calculating a correction coefficient of each lamp point in the LED display screen according to the corrected image after dead pixels are removed, and sending the correction coefficient to the sending card;
the sending card is connected with the receiving card of the LED display screen and used for sending the correction coefficient to the receiving card of the LED display screen;
and the LED display screen is used for correcting the display data through the receiving card according to the received correction coefficient of each lamp point and displaying the corrected display data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210602049.9A CN115018936A (en) | 2022-05-30 | 2022-05-30 | Image dead pixel removing method, device and correction system based on normal distribution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210602049.9A CN115018936A (en) | 2022-05-30 | 2022-05-30 | Image dead pixel removing method, device and correction system based on normal distribution |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115018936A true CN115018936A (en) | 2022-09-06 |
Family
ID=83070999
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210602049.9A Pending CN115018936A (en) | 2022-05-30 | 2022-05-30 | Image dead pixel removing method, device and correction system based on normal distribution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115018936A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116363006A (en) * | 2023-03-28 | 2023-06-30 | 北京拙河科技有限公司 | Image calibration method and device based on normal algorithm |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6014471A (en) * | 1996-09-08 | 2000-01-11 | Scitex Corporation | Apparatus and method for retouching a digital representation of a color image |
| CN101588505A (en) * | 2008-05-21 | 2009-11-25 | 英华达（南京）科技有限公司 | The digital photographing apparatus of the bad point of the method for correcting dead pixel on digital camera device point and tool zero offset capability |
| KR20110079317A (en) * | 2009-12-31 | 2011-07-07 | 주식회사 동부하이텍 | Method of detection and correction of bad pixel |
| US20140133774A1 (en) * | 2012-11-09 | 2014-05-15 | Industrial Technology Research Institute | Image processor and image dead pixel detection method thereof |
| CN104504722A (en) * | 2015-01-09 | 2015-04-08 | 电子科技大学 | Method for correcting image colors through gray points |
| CN106210712A (en) * | 2016-08-11 | 2016-12-07 | 上海大学 | A kind of dead pixel points of images detection and processing method |
| WO2020181522A1 (en) * | 2019-03-13 | 2020-09-17 | 深圳市汇顶科技股份有限公司 | Defective pixel detection method, image processing chip, and electronic device |
| CN112019775A (en) * | 2020-09-04 | 2020-12-01 | 成都微光集电科技有限公司 | Bad point detection and correction method and device |
| CN112911174A (en) * | 2021-01-18 | 2021-06-04 | 珠海全志科技股份有限公司 | Image dead pixel cluster correction method, computer device and computer readable storage medium |
| CN114245043A (en) * | 2021-12-16 | 2022-03-25 | 广东赛昉科技有限公司 | Image dead pixel dynamic correction and ASIC implementation method and system thereof |
-
2022
- 2022-05-30 CN CN202210602049.9A patent/CN115018936A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6014471A (en) * | 1996-09-08 | 2000-01-11 | Scitex Corporation | Apparatus and method for retouching a digital representation of a color image |
| CN101588505A (en) * | 2008-05-21 | 2009-11-25 | 英华达（南京）科技有限公司 | The digital photographing apparatus of the bad point of the method for correcting dead pixel on digital camera device point and tool zero offset capability |
| KR20110079317A (en) * | 2009-12-31 | 2011-07-07 | 주식회사 동부하이텍 | Method of detection and correction of bad pixel |
| US20140133774A1 (en) * | 2012-11-09 | 2014-05-15 | Industrial Technology Research Institute | Image processor and image dead pixel detection method thereof |
| CN104504722A (en) * | 2015-01-09 | 2015-04-08 | 电子科技大学 | Method for correcting image colors through gray points |
| CN106210712A (en) * | 2016-08-11 | 2016-12-07 | 上海大学 | A kind of dead pixel points of images detection and processing method |
| WO2020181522A1 (en) * | 2019-03-13 | 2020-09-17 | 深圳市汇顶科技股份有限公司 | Defective pixel detection method, image processing chip, and electronic device |
| CN112019775A (en) * | 2020-09-04 | 2020-12-01 | 成都微光集电科技有限公司 | Bad point detection and correction method and device |
| CN112911174A (en) * | 2021-01-18 | 2021-06-04 | 珠海全志科技股份有限公司 | Image dead pixel cluster correction method, computer device and computer readable storage medium |
| CN114245043A (en) * | 2021-12-16 | 2022-03-25 | 广东赛昉科技有限公司 | Image dead pixel dynamic correction and ASIC implementation method and system thereof |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116363006A (en) * | 2023-03-28 | 2023-06-30 | 北京拙河科技有限公司 | Image calibration method and device based on normal algorithm |
| CN116363006B (en) * | 2023-03-28 | 2024-02-02 | 北京拙河科技有限公司 | Image calibration method and device based on normal algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100491952C (en) | Image inspection device, image inspection method, and image inspection program | |
| CN109741307B (en) | Stray light detection method, stray light detection device and stray light detection system of camera module | |
| CN110954553B (en) | Lens optical detection system | |
| CN111008563B (en) | Dim light scene seed germination detection method and device and readable storage medium | |
| CN111210764B (en) | LED screen correction method and correction device | |
| CN115018936A (en) | Image dead pixel removing method, device and correction system based on normal distribution | |
| CN108805144A (en) | Shell hole recognition methods based on morphology correction and system, indication of shots equipment | |
| CN112102762A (en) | Method and device for identifying abnormal light points of LED screen | |
| CN113808135B (en) | Image brightness abnormality detection method, electronic device, and storage medium | |
| CN117274211A (en) | Screen defect detection method and device, terminal equipment and storage medium | |
| JP2928714B2 (en) | Cell activity determination method and device | |
| CN116016909A (en) | Television partition backlight detection system and method | |
| CN106898084B (en) | method and device for determining image processing threshold | |
| CN114387182A (en) | Camera module lens shadow correction index standard exceeding detection method and system | |
| CN112785963A (en) | LED lamp bead positioning method and device and storage medium | |
| CN113012137B (en) | Panel defect inspection method, system, terminal device and storage medium | |
| JP7177753B2 (en) | Material separation evaluation method and material separation evaluation system | |
| CN103226689A (en) | Red eye detection method and device and red eye removing method and device | |
| CN112183158A (en) | Grain type identification method of grain cooking equipment and grain cooking equipment | |
| CN114088000B (en) | Molten steel liquid level distance determining method, system, equipment and medium | |
| CN117830319B (en) | Power adapter product detection method based on image processing | |
| JPH08247962A (en) | Defect detection method and apparatus for color filter and defect detection method and apparatus for panel display | |
| CN114078426B (en) | Method for eliminating influence of camera curve on lamp point image and display screen correction method | |
| CN117522866B (en) | Method for judging silk thread anchor points in fluorescent microfilament test image based on mask | |
| CN115713501B (en) | Detection processing method and system suitable for blurred pictures of camera |
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 |