A method, device, medium and equipment for detecting laser dead pixels

Abstract

The present invention provides a method, device, medium and equipment for detecting laser dead pixels. The method includes: acquiring a luminous point image of a transmitting laser, determining target pixels in the luminous point image, and obtaining a to-be-detected image according to the target pixels; creating a Gaussian convolution kernel model, performing convolution processing on the to-be-detected image based on the Gaussian convolution model to obtain a feature map; performing non-maximum suppression processing on the feature map to determine a plurality of reference local maximum values; screening the plurality of reference local maximum values to obtain a target maximum value; the pixel point corresponding to the target maximum value is a dead pixel; thus, regardless of the model and type of the transmitting laser, as long as the luminous point image of the transmitting laser is acquired, the Gaussian convolution kernel model can be used to determine the dead pixel; this testing method has good versatility and can improve the testing efficiency.

Description

Technical Field

[0001] This application relates to the technical field of laser optical property testing, and particularly to a method, device, medium, and equipment for detecting laser dead pixels. Background Art

[0002] Currently, the Time-of-Flight (TOF) module has been widely used in depth distance testing. The TOF (Time of Flight) technology is a 3D imaging technology that emits measurement light from a transmitter, reflects it back to the receiver through a target object, and calculates the distance from the object to the receiving sensor based on the time taken by the measurement light to travel this distance. To ensure the test accuracy of the TOF module, it is necessary to evaluate the performance of the emitting laser and determine whether the performance of the emitting laser meets the requirements.

[0003] The main test item of the emitting laser is dead pixel testing. Different manufacturers have different test schemes. For the emitting lasers of different manufacturers, R & D personnel need to learn the corresponding test methods, which takes a long time.

[0004] It can be seen that when testing for dead pixels of the emitting laser in the prior art, the test method lacks generality, resulting in a reduction in test efficiency. Summary of the Invention

[0005] In view of the problems existing in the prior art, embodiments of the present invention provide a method, device, medium, and equipment for detecting laser dead pixels to solve or partially solve the technical problem that when testing for dead pixels of a laser in the prior art, the test method lacks generality, resulting in a reduction in test efficiency.

[0006] In the first aspect of the present invention, a method for detecting laser dead pixels is provided, characterized in that the method includes:

[0007] Obtain the light-emitting point image of the emitting laser, determine the target pixels in the light-emitting point image, and obtain the image to be detected based on the target pixels;

[0008] Create a Gaussian convolution kernel model, and perform convolution processing on the image to be detected based on the Gaussian convolution model to obtain a feature map;

[0009] Perform non-maximum suppression processing on the feature map to determine a plurality of reference local maxima;

[0010] Screen the plurality of reference local maxima to obtain the target maximum; the pixel point corresponding to the target maximum is the dead pixel.

[0011] In the above solution, the determining the target pixels in the light-emitting point image includes:

[0012] Determine the image background threshold based on the luminous point image; the image background threshold is the mean value of the gray values of all pixel points in the luminous point image;

[0013] Take the pixel points with gray values greater than the image background threshold as target pixels.

[0014] In the above solution, the creation of the Gaussian convolution kernel model includes:

[0015] Determine the blur radius of the Gaussian convolution kernel model;

[0016] Determine the initial pixel matrix based on the blur radius;

[0017] Using the formula Determine the weight value corresponding to each pixel point in the initial pixel matrix;

[0018] Fill each of the weight values into the corresponding position in the initial pixel matrix to form the Gaussian convolution kernel model; where

[0019] The x is the coordinate of each pixel point in the initial pixel matrix, the y is the weight value corresponding to each pixel point, μ is the mean value of the Gaussian convolution kernel, and σ is the standard deviation of the Gaussian convolution kernel.

[0020] In the above solution, after creating the Gaussian convolution kernel model, the method further includes:

[0021] Obtain the smoothness of the feature map, and adjust the blur radius of the Gaussian convolution kernel model according to the smoothness.

[0022] In the above solution, the non-maximum suppression processing of the feature map to determine multiple reference local maxima includes:

[0023] Traverse the feature map with a preset first local window, and take the maximum gray value of each pixel point in each first local window as the initial local maximum;

[0024] With each of the initial local maxima as the center, determine the corresponding target detection image based on a preset second local window;

[0025] Take the maximum gray value of each pixel point in the target detection image as the reference local maximum.

[0026] In the above solution, the screening of the multiple reference local maxima to obtain the target maximum includes:

[0027] Obtain the gray value of the pixel point corresponding to each reference local maximum;

[0028] Determine the gray mean value of the gray values of the pixel points corresponding to each reference local maximum;

[0029] Determine a grayscale value threshold based on the average grayscale value;

[0030] Determine the reference local maximum with a pixel grayscale value less than the grayscale value threshold as the target maximum.

[0031] In a second aspect of the present invention, there is provided a device for detecting laser dead pixels, the device comprising:

[0032] A determination unit configured to obtain an image of the light-emitting points of a transmitting laser, determine target pixels in the image of the light-emitting points, and obtain a to-be-detected image based on the target pixels;

[0033] A creation unit configured to create a Gaussian convolution kernel model, perform convolution processing on the to-be-detected image based on the Gaussian convolution model, and obtain a feature map;

[0034] A suppression unit configured to perform non-maximum suppression processing on the feature map to determine a plurality of reference local maxima;

[0035] A screening unit configured to screen the plurality of reference local maxima to obtain a target maximum; the pixel point corresponding to the target maximum is a dead pixel.

[0036] In the above solution, the determination unit is specifically configured to:

[0037] Determine an image background threshold based on the image of the light-emitting points; the image background threshold is the average value of the grayscale values of all pixel points in the image of the light-emitting points;

[0038] Use the pixel points with grayscale values greater than the image background threshold as target pixels.

[0039] In a third aspect of the present invention, there is provided a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of the method according to any one of the first aspects are implemented.

[0040] In a fourth aspect of the present invention, there is provided a computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the steps of the method according to any one of the first aspects are implemented.

[0041] The present invention provides a method, apparatus, medium, and device for detecting laser dead pixels. The method includes: obtaining a light-emitting point image of a transmitting laser, determining target pixels in the light-emitting point image, and obtaining a to-be-detected image based on the target pixels; creating a Gaussian convolution kernel model, performing convolution processing on the to-be-detected image based on the Gaussian convolution model to obtain a feature map; performing non-maximum suppression processing on the feature map to determine a plurality of reference local maximum values; screening the plurality of reference local maximum values to obtain a target maximum value; the pixel point corresponding to the target maximum value is a dead pixel; thus, regardless of the model and type of the transmitting laser, as long as the light-emitting point image of the transmitting laser is obtained, the dead pixel can be determined by using the Gaussian convolution kernel model; this testing method has good versatility and can improve the testing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components.

[0043] In the drawings:

[0044] Figure 1 shows a schematic flowchart of a method for detecting laser dead pixels according to an embodiment of the present invention;

[0045] Figure 2 shows a schematic diagram of a to-be-detected image according to an embodiment of the present invention;

[0046] Figure 3 shows a schematic diagram of a Gaussian convolution kernel model according to an embodiment of the present invention;

[0047] Figure 4 shows a schematic diagram of a feature map according to an embodiment of the present invention;

[0048] Figure 5 shows a schematic diagram of an image obtained after performing non-maximum suppression processing on a feature map according to an embodiment of the present invention;

[0049] Figure 6 shows a schematic diagram of the determined dead pixel position according to an embodiment of the present invention;

[0050] Figure 7 shows a schematic structural diagram of an apparatus for detecting laser dead pixels according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art.

[0052] This embodiment provides a method for detecting laser dead pixels, as Figure 1 shown, the method includes the following steps:

[0053] S110, obtain the light-emitting point image of the emitting laser, determine the target pixels in the light-emitting point image, and obtain the image to be detected according to the target pixels;

[0054] In this embodiment, a camera (such as an infrared CCD camera) can be used in combination with a telephoto lens to obtain the light-emitting point image of the emitting laser. Specifically, the exposure of the camera can be adjusted until each divergence point of the emitting laser is clearly imaged, and then a VCSEL light-emitting point image is taken. Among them, the emitting laser can be a vertical cavity surface emitting laser (VCSEL, Vertical-Cavity Surface-Emitting Laser).

[0055] Then determine the target pixels in the light-emitting point image, and obtain the image to be detected according to the target pixels; among them, the image to be detected is an image composed of the target pixels, and the image to be detected can be as Figure 2 shown.

[0056] In one implementation, determining the target pixels in the light-emitting point image includes:

[0057] Determine the image background threshold based on the light-emitting point image; the image background threshold is the mean value of the gray values of all pixel points in the light-emitting point image;

[0058] Take the pixel points with gray values greater than the image background threshold as the target pixels.

[0059] For example, assume that the resolution of the light-emitting point image is 2048*2048, count the gray values of all pixel points within the entire image resolution range, and determine the mean value of the gray values of all pixel points, and this mean value is the image background threshold.

[0060] S111, create a Gaussian convolution kernel model, and perform convolution processing on the image to be detected based on the Gaussian convolution model to obtain a feature map;

[0061] In this embodiment, the Gaussian convolution kernel model is used to perform convolution on the image to be detected. The pixel points at the positions of the dead pixels in the convolved image will have Gaussian distribution characteristics, so that the dead pixels can be accurately determined and the determination accuracy of the dead pixels can be improved.

[0062] Then, it is necessary to first create a Gaussian convolution kernel model, including:

[0063] Determine the blur radius of the Gaussian convolution kernel model;

[0064] Determine the initial pixel matrix based on the blur radius;

[0065] Use the Gaussian function Determine the weight value corresponding to each pixel point in the initial pixel matrix;

[0066] Fill each weight value into the corresponding position in the initial pixel matrix to form a Gaussian convolution kernel model; where,

[0067] x is the coordinate of each pixel point in the initial pixel matrix, y is the weight value corresponding to each pixel point, μ is the mean of the Gaussian convolution kernel, and σ is the standard deviation of the Gaussian convolution kernel.

[0068] Specifically, the Gaussian convolution kernel model is essentially a weight matrix. The process of performing convolution processing on the image to be detected based on the Gaussian convolution kernel model is essentially a process of multiplying the Gaussian convolution kernel model by the two-dimensional matrix corresponding to the image to be detected. After multiplication, a feature map can be obtained. Therefore, the accuracy of the Gaussian convolution kernel model determines the processing effect of the image. Therefore, in this embodiment, it is necessary to first determine the blur radius of the Gaussian convolution kernel model.

[0069] The blur radius is the distance value between the center point of the Gaussian convolution kernel model and the surrounding pixel points. Refer to Table 1. Assuming the center point is (0,0) and the blur radius is 2, then the initial pixel matrix is a pixel matrix formed by expanding 2 rows / columns of pixels upward, downward, leftward, and rightward centered on the center point. The initial pixel matrix is also a 5*5 order matrix. Among them, the data in each position in Table 1 is the coordinate of the pixel.

[0070] Table 1

[0071] (-2,2) (-1,2) (0,2) (1,2) (2,2) (-2,1) (-1,1) (0,1) (1,1) (2,1) (-2,0) (-1,0) (0,0) (1,0) (2,0) (-2,-1) (-1,-1) (0,-1) (1,-1) (2,-1) (-2,-2) (-1,-2) (0,-2) (1,-2) (2,-2)

[0072] Then set the standard deviation and mean of the Gaussian convolution kernel, and use the Gaussian function to calculate the weight value corresponding to each pixel point in the initial pixel matrix. For example, the mean can be set to 0 and the standard deviation can be set to 3; it can also be set based on the actual situation and is not limited here.

[0073] The matrix formed by the weight values corresponding to each pixel point is the Gaussian convolution kernel model.

[0074] It should be noted that the initially determined blur radius may not meet the image processing requirements. Therefore, after creating the Gaussian convolution kernel model, the method further includes:

[0075] Obtain the smoothness of the feature map, and adjust the blur radius of the Gaussian convolution kernel model according to the smoothness.

[0076] For example, when the smoothness of the feature map is lower than the smoothness threshold, the blur radius needs to be increased; if the smoothness is greater than the smoothness threshold, the blur radius needs to be decreased. Each time of increasing or decreasing, it can be increased or decreased by a unit blur radius, and the unit blur radius is 1. Among them, the smoothness threshold can be set based on the actual situation and is not limited here.

[0077] In this embodiment, the Gaussian convolution kernel model can refer to Figure 3 After convolving the image to be detected using the Gaussian convolution kernel model, the obtained feature map can refer to Figure 4

[0078] S112, perform non-maximum suppression processing on the feature map to determine multiple reference local maxima;

[0079] After determining the feature map, it is necessary to perform non-maximum suppression processing on the feature map to determine multiple reference local maxima, including:

[0080] Traverse the feature map with a preset first local window, and use the maximum gray value of each pixel point in each first local window as the initial local maximum;

[0081] With each initial local maximum as the center, determine the corresponding target detection image based on a preset second local window;

[0082] Use the maximum gray value of each pixel point in the target detection image as the reference local maximum.

[0083] Specifically, the first local window can be 10*10, which is equivalent to traversing 100 pixel points each time; then obtain the maximum gray value in each first local window, and use the maximum gray value in each first local window as the initial local maximum. Among them, there are no overlapping pixel points in two adjacent first local windows.

[0084] Then, assign the initial local maximum to the original gray value of the pixel point corresponding to the initial local maximum, and assign non-initial local maxima to 0 (it can be understood that the gray value of the pixel point corresponding to the initial local non-maximum is set to 0). This can highlight the brightness and position of the pixel point corresponding to the initial local maximum.

[0085] ​Secondly, centering on each initial local maximum, all pixel points within the second local window are selected as the image to be detected. The size of the second local window can be the same as or different from that of the first local window. For example, the size of the second local window can be 15*15, that is, 225 pixel points are screened. In this way, since the size of the second local window is larger than that of the first local window, the second layout window is very likely to contain multiple initial local maxima. Therefore, it is equivalent to screening multiple initial local maxima again to ensure the accuracy of the local maxima.

[0086] Similarly, the maximum gray value in the second local window is used as the reference local maximum. The pixel point corresponding to the reference local maximum is assigned the original gray value of the pixel point corresponding to the reference local maximum, and the pixel points in the second local window that do not correspond to the reference local maximum are assigned 0. In this way, the non-maximum suppression operation is completed, which can highlight the brightness and position of the pixel point corresponding to the reference local maximum, facilitating subsequent positioning of bad points based on the local maximum. Among them, the local maximum image obtained after performing non-maximum suppression processing on the feature map can be referred to Figure 5 .

[0087] For example, if the second local window contains the initial local maximum A and the initial local maximum B, and the gray value corresponding to A is greater than the gray value corresponding to B, then the initial local maximum A is used as the reference local maximum and assigned 1; and the initial local maximum B is assigned 0.

[0088] It should be noted that if there are two initial local maxima with the same gray value in the second local window (this situation generally does not occur), both initial local maxima with the same gray value need to be retained.

[0089] S113. Screen the multiple reference local maxima to obtain the target maximum; the pixel point corresponding to the target maximum is a bad point.

[0090] After determining the local maximum, it does not mean that all pixel points corresponding to the local maximum are bad points. It is still necessary to screen the multiple reference local maxima to obtain the target maximum, and then determine the bad points based on the target maximum.

[0091] In one implementation, screening the multiple reference local maxima to obtain the target maximum includes:

[0092] Obtain the gray value of the pixel point corresponding to each reference local maximum;

[0093] Determine the gray value average of the gray value of the pixel point corresponding to each reference local maximum;

[0094] Determine the gray value threshold based on the gray value average;

[0095] Determine the reference local maximum with a pixel grayscale value less than the grayscale value threshold as the target maximum. Among them, half of the grayscale mean is the grayscale value threshold.

[0096] Finally, determine the pixel corresponding to the target maximum as a bad pixel, and the position of the pixel corresponding to the target maximum is the bad pixel position; the determined bad pixel map can be referred to Figure 6 , in Figure 6 the bad pixel positions are (365, 53) and (299, 572).

[0097] In this way, no matter what type and kind of emission laser, as long as the light-emitting point image of the emission laser is obtained, the bad pixels can be determined by using the Gaussian convolution kernel model; this test method has good versatility and can improve the test efficiency.

[0098] Based on the same inventive concept as the foregoing embodiment, this embodiment also provides a device for detecting bad pixels of a VCSEL, as Figure 7 shown, the device includes:

[0099] A determination unit 71, configured to obtain a light-emitting point image of an emission laser, determine target pixels in the light-emitting point image, and obtain a to-be-detected image according to the target pixels;

[0100] A creation unit 72, configured to create a Gaussian convolution kernel model, perform convolution processing on the to-be-detected image based on the Gaussian convolution model, and obtain a feature map;

[0101] A suppression unit 73, configured to perform non-maximum suppression processing on the feature map to determine a plurality of reference local maxima;

[0102] A screening unit 74, configured to screen the plurality of reference local maxima to obtain a target maximum; the pixel corresponding to the target maximum is a bad pixel.

[0103] In one implementation manner, the determination unit 71 is specifically configured to:

[0104] Determine an image background threshold based on the light-emitting point image; the image background threshold is the mean of the grayscale values of all pixel points in the light-emitting point image;

[0105] Take the pixel points with grayscale values greater than the image background threshold as target pixels.

[0106] Since the device introduced in the embodiments of the present invention is the device adopted for implementing the method for detecting laser dead pixels in the embodiments of the present invention, based on the method introduced in the embodiments of the present invention, those skilled in the art can understand the specific structure and variations of the device, so it will not be elaborated here. Any device adopted by the method of the embodiments of the present invention falls within the scope of protection of the present invention.

[0107] Based on the same inventive concept as in the foregoing embodiments, the embodiments of the present invention also provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of any of the foregoing methods are implemented.

[0108] Based on the same inventive concept as in the foregoing embodiments, the embodiments of the present invention also provide a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of any of the foregoing methods are implemented.

[0109] Through one or more embodiments of the present invention, the present invention has the following beneficial effects or advantages:

[0110] The present invention provides a method, device, medium, and device for detecting laser dead pixels. The method includes: obtaining a light-emitting point image of a transmitting laser, determining target pixels in the light-emitting point image, and obtaining a to-be-detected image according to the target pixels; creating a Gaussian convolution kernel model, performing convolution processing on the to-be-detected image based on the Gaussian convolution model to obtain a feature map; performing non-maximum suppression processing on the feature map to determine a plurality of reference local maximum values; screening the plurality of reference local maximum values to obtain a target maximum value; the pixel point corresponding to the target maximum value is a dead pixel; thus, no matter what model and type of transmitting laser it is, as long as the light-emitting point image of the transmitting laser is obtained, the dead pixel can be determined by using the Gaussian convolution kernel model; this test method has good versatility and can improve the test efficiency.

[0111] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The structure required to construct such systems will be apparent from the above description. In addition, the present invention is not directed to any particular programming language. It should be understood that the content of the present invention described herein can be implemented using various programming languages, and the description of the specific language above is for disclosing the best mode of the present invention.

[0112] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that the embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures, and technologies have not been shown in detail so as not to obscure the understanding of this specification.

[0113] Similarly, it should be understood that, for the purpose of streamlining the present disclosure and assisting in understanding one or more of the various inventive aspects, in the foregoing description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the following claims, the inventive aspects lie in less than all the features of the single embodiments disclosed previously. Thus, the claims following the detailed description hereby expressly incorporate the detailed description, where each claim itself serves as a separate embodiment of the present invention.

[0114] Those skilled in the art will appreciate that the modules in the devices in the embodiments can be adaptively changed and disposed in one or more devices different from those of the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Any combination can be adopted for all the features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all the processes or units of any method or device so disclosed, unless at least some of such features and / or processes or units are mutually exclusive. Each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) can be replaced by an alternative feature that provides the same, equivalent, or similar purpose, unless otherwise expressly stated.

[0115] In addition, those skilled in the art will be able to understand that, although some of the embodiments herein include certain features included in other embodiments rather than other features, the combination of the features of different embodiments means that it is within the scope of the present invention and forms different embodiments. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

[0116] Each component embodiment of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components in the gateway, proxy server, and system according to the embodiments of the present invention. The present invention can also be implemented as a device or device program (such as a computer program and a computer program product) for executing part or all of the methods described herein. Such a program for implementing the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0117] It should be noted that the above embodiments illustrate the present invention rather than limit the present invention, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in the claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention can be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In a unit claim listing several devices, several of these devices can be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names.

[0118] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0119] As described above, it is only the preferred embodiments of the present invention, and is not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for detecting laser dead pixels, characterized in that, The method includes: Obtaining an image of the light-emitting point of the emission laser, determining target pixels in the image of the light-emitting point, and obtaining a to-be-detected image based on the target pixels; Creating a Gaussian convolution kernel model, performing convolution processing on the to-be-detected image based on the Gaussian convolution model, and obtaining a feature map; Performing non-maximum suppression processing on the feature map to determine a plurality of reference local maxima; Screening the plurality of reference local maxima to obtain a target maximum; the pixel point corresponding to the target maximum is a defective pixel; wherein, The performing non-maximum suppression processing on the feature map to determine a plurality of reference local maxima includes: Traversing the feature map with a preset first local window, and taking the maximum gray value of each pixel point in each first local window as an initial local maximum; Taking each of the initial local maxima as a center, and determining a corresponding target detection image based on a preset second local window; Taking the maximum gray value of each pixel point in the target detection image as the reference local maximum.

2. The method according to claim 1, characterized in that, The determining the target pixels in the image of the light-emitting point includes: Determining an image background threshold based on the image of the light-emitting point; the image background threshold is the mean value of the gray values of all pixel points in the image of the light-emitting point; Taking the pixel points with gray values greater than the image background threshold as target pixels.

3. The method according to claim 1, wherein The creating the Gaussian convolution kernel model includes: Determining the blur radius of the Gaussian convolution kernel model; Determining an initial pixel matrix based on the blur radius; Using the formula to determine the weight value corresponding to each pixel point in the initial pixel matrix; Filling each of the weight values into the corresponding position in the initial pixel matrix to form the Gaussian convolution kernel model; wherein, x is the coordinate of each pixel point in the initial pixel matrix, y is the corresponding weight value of each pixel point, μ is the mean value of the Gaussian convolution kernel, and σ is the standard deviation of the Gaussian convolution kernel.

4. The method according to claim 3, wherein After creating the Gaussian convolution kernel model, the method further includes: Obtaining the smoothness of the feature map, and adjusting the blur radius of the Gaussian convolution kernel model according to the smoothness.

5. The method according to claim 1, wherein The screening the plurality of reference local maxima to obtain a target maximum includes: Obtaining the gray value of the pixel point corresponding to each reference local maximum; Determining the gray mean value of the gray values of the pixel points corresponding to all reference local maxima; Determining a gray value threshold based on the gray mean value; Determining the reference local maxima with pixel point gray values less than the gray value threshold as the target maximum.

6. A device for detecting laser dead pixels, characterized in that, The device includes: A determination unit, configured to obtain an image of the light-emitting point of the emission laser, determine target pixels in the image of the light-emitting point, and obtain a to-be-detected image based on the target pixels; A creation unit, configured to create a Gaussian convolution kernel model, perform convolution processing on the to-be-detected image based on the Gaussian convolution model, and obtain a feature map; A suppression unit, configured to perform non-maximum suppression processing on the feature map to determine a plurality of reference local maxima; A screening unit, configured to screen the plurality of reference local maxima to obtain a target maximum; the pixel point corresponding to the target maximum is a defective pixel; wherein, The performing non-maximum suppression processing on the feature map to determine a plurality of reference local maxima includes: Traverse the feature map with a preset first local window, and use the maximum gray value of each pixel point in each first local window as the initial local maximum value; With each of the initial local maximum values as the center, determine the corresponding target detection image based on a preset second local window; Use the maximum gray value of each pixel point in the target detection image as the reference local maximum value.

7. The device according to claim 6, characterized in that, The determining unit is specifically configured to: Determine an image background threshold based on the light-emitting point image; the image background threshold is the mean value of the gray values of all pixel points in the light-emitting point image; Use the pixel points with gray values greater than the image background threshold as target pixels.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the steps of the method according to any one of claims 1-5.

9. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method according to any one of claims 1-5.

Images