Method and device for correcting image transmitted by vehicle-mounted lens, and medium

Abstract

The invention provides a correction method and device for a transmission image of a vehicle-mounted lens, and a medium, and relates to the technical field of image processing, and the method comprises the steps: taking a transmission image shot by the vehicle-mounted lens, obtaining the target speed information of the vehicle-mounted lens in the exposure time of shooting the transmission image, and obtaining N, N ', N' ', N' ', N' 'and N' '; the method comprises the steps of acquiring target speed information, uniformly dividing exposure time into N target time slices, acquiring a target vehicle-mounted lens pose corresponding to each target time slice based on the target speed information, determining a target point diffusion matrix based on the target vehicle-mounted lens pose corresponding to each target time slice, and processing a transmission image by using the target point diffusion matrix to obtain a target vehicle-mounted lens pose corresponding to each target time slice. A target clear image is obtained, and the calculation efficiency and the definition of the transmitted image are balanced.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method, device and medium for correcting images transmitted by a vehicle-mounted lens. Background Art

[0002] In autonomous driving, in-vehicle navigation and assisted driving systems, the quality of images captured by on-board cameras directly affects the accuracy of key tasks such as target detection and environmental perception. However, during vehicle driving, since the vehicle may travel on various routes such as straight lines and arcs, the on-board camera experiences rapid acceleration changes and angular velocity changes during the exposure time, resulting in nonlinear characteristics in the deblurred point spread function of the image, and traditional deblurring methods face challenges. In the prior art, the global deblurring algorithm based on uniform motion is difficult to accurately model the point spread matrix of the vehicle, and dividing the exposure time into a fixed number of time slices ignores the dynamic changes in motion intensity. Patent document No. 202110280229.5 discloses a system and method for image deblurring in a vehicle, but the patent document only discloses the use of velocity and acceleration information to obtain the point spread matrix, and does not involve angular velocity. Therefore, an adaptive time slice division mechanism based on comprehensive velocity is still needed to achieve a balance between efficiency and clear image accuracy. Summary of the Invention

[0003] In view of the above technical problems, the technical solution adopted by the present invention is:

[0004] According to a first aspect of the present invention, a method for correcting an image transmitted by a vehicle-mounted lens is provided, the method comprising the following steps:

[0005] Acquire a transmission image captured by the vehicle-mounted lens, and acquire target velocity information of the vehicle-mounted lens during an exposure time for capturing the transmission image, wherein the target velocity information includes at least a target linear acceleration and a target angular velocity;

[0006] Obtain N, and evenly divide the exposure time into N target time slices, and obtain the target vehicle-mounted lens pose corresponding to each target time slice based on the target speed information;

[0007] Based on the target vehicle lens pose corresponding to each target time slice, the target point spread matrix is ​​determined, and the target point spread matrix is ​​used to process the transmitted image to obtain a clear target image;

[0008] Among them, N is obtained through the following steps:

[0009] Determine the target integrated velocity cv based on the target linear acceleration and the target angular velocity;

[0010] If cv is greater than v0, obtain N, where N satisfies the following condition: N is equal to the average of all time slice numbers greater than N0 in the time slice number list, where the time slice number list includes several time slice numbers, v0 is the speed threshold, and N0 is the time slice number threshold;

[0011] If cv is not greater than v0, obtain N, where N satisfies the following condition: N is equal to the average of all time slice numbers not greater than N0 in the time slice number list.

[0012] According to a second aspect of the present invention, a non-transitory computer-readable storage medium is provided, in which a computer program is stored. The computer program is loaded and executed by a processor to implement the aforementioned method.

[0013] According to a third aspect of the present invention, an electronic device is provided, comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the aforementioned method when executing the computer program.

[0014] The present invention has at least the following beneficial effects: obtaining a transmission image captured by a vehicle-mounted lens, and obtaining target speed information of the vehicle-mounted lens during the exposure time for capturing the transmission image, obtaining N, and evenly dividing the exposure time into N target time slices, and obtaining a target vehicle-mounted lens posture corresponding to each target time slice based on the target speed information, determining a target point spread matrix based on the target vehicle-mounted lens posture corresponding to each target time slice, and using the target point spread matrix to process the transmission image to obtain a clear target image. The present invention dynamically optimizes the segmentation granularity within the exposure time by means of a speed threshold and a time slice number threshold, thereby achieving a balance between computational efficiency and the clarity of the transmission image. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A flowchart of a method for correcting images transmitted by a vehicle-mounted lens provided in an embodiment of the present invention;

[0017] Figure 2 A flowchart of S200 provided in an embodiment of the present invention;

[0018] Figure 3 A flowchart of S220 provided in an embodiment of the present invention;

[0019] Figure 4 A flowchart of an embodiment of S227 provided in an embodiment of the present invention;

[0020] Figure 5 A flowchart of another embodiment of S227 provided in an embodiment of the present invention;

[0021] Figure 6 A flowchart of S300 provided in an embodiment of the present invention;

[0022] Figure 7 A flowchart of S320 provided in an embodiment of the present invention;

[0023] Figure 8 This is a flowchart of another embodiment of S220 provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar tasks and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] The embodiment of the present invention provides a method for correcting images transmitted by a vehicle-mounted lens, such as Figure 1 As shown, the method includes the following steps:

[0027] S100: Acquire a transmission image captured by a vehicle-mounted camera lens, and obtain target velocity information during the exposure time of the vehicle-mounted camera lens during the transmission image capture. The target velocity information includes at least a target linear acceleration and a target angular velocity. Specifically, the target velocity information also includes vehicle velocity. Furthermore, the target angular velocity and target linear acceleration are acquired using an onboard IMU. Specifically, based on the IMU acquisition frequency, the target velocity information includes the target linear acceleration and target angular velocity acquired by the IMU each time during the exposure time.

[0028] S200, obtain N, and evenly divide the exposure time into N target time slices, and obtain the target vehicle lens posture corresponding to each target time slice based on the target speed information. Specifically, for each target time slice t i , the rotation matrix R(t i ), the displacement T(t i ), thereby obtaining the target vehicle-mounted lens pose [R(t i ), T(t i )].

[0029] S300: Determine a target point spread matrix based on the target vehicle-mounted lens posture corresponding to each target time slice, and use the target point spread matrix to process the transmitted image to obtain a clear target image.

[0030] Among them, such as Figure 2 As shown, in S200, N is obtained through the following steps:

[0031] S210 : Determine a target integrated velocity cv based on the target linear acceleration and the target angular velocity.

[0032] In one embodiment of the present invention, the target integrated velocity is an average value of the first sum value collected each time by the IMU, and the first sum value is equal to the sum of the target angular velocity and the target linear acceleration.

[0033] In another embodiment of the present invention, the target is integrated as the average value of the second sum value collected by the IMU each time, and the second sum value is equal to v cm +w×b,v cm is the sum of the integral of the target linear acceleration in one IMU acquisition time interval and the initial velocity, w is the target angular velocity, and b is the position vector of the on-board lens relative to the center of mass of the vehicle.

[0034] S220, if cv is greater than v0, obtain N, where N satisfies the following condition: N is equal to the average of all time slice numbers greater than N0 in the time slice number list, wherein the time slice number list includes a plurality of time slice numbers, v0 is the speed threshold, and N0 is the time slice number threshold. Specifically, the time slice list A = {A1, A2, ..., A r ,…,A s}, A r+1 Greater than A r , if A r Not less than N0, and A r-1 Less than N0, get A r to A s The average value of is taken as N.

[0035] S230, if cv is not greater than v0, obtain N, where N satisfies the following condition: N is equal to the average of all time slice numbers not greater than N0 in the time slice number list. Specifically, if A r+1 Greater than N0, and A r Not greater than N0, get A1 to A s The average value of is taken as N.

[0036] In summary, the transmission image taken by the vehicle-mounted lens is obtained, and the target speed information of the vehicle-mounted lens during the exposure time of shooting the transmission image is obtained. N is obtained, and the exposure time is evenly divided into N target time slices. The target vehicle-mounted lens posture corresponding to each target time slice is obtained based on the target speed information. Based on the target vehicle-mounted lens posture corresponding to each target time slice, the target point spread matrix is ​​determined, and the transmission image is processed using the target point spread matrix to obtain a clear target image. The present invention dynamically optimizes the segmentation granularity within the exposure time by means of a speed threshold and a time slice number threshold, thereby achieving a balance between computational efficiency and the clarity of the transmission image.

[0037] Specifically, such as Figure 3 As shown, in S220, v0 and N0 are obtained through the following steps:

[0038] S221, obtain sample comprehensive speed list v={v1, v2, ..., v j ,…,v n}, v j is the jth sample integrated speed, where j ranges from 1 to n, and n is the number of sample integrated speeds. The sample integrated speed is determined based on the sample speed information of the sample vehicle during the exposure time of one sample, and the sample speed information includes at least the sample linear acceleration and the sample angular velocity.

[0039] Specifically, the sample speed information includes the sample linear acceleration and sample angular velocity collected each time by the IMU of the sample vehicle during the exposure time of the sample. In one embodiment of the present invention, the sample linear acceleration and sample angular velocity collected each time during the exposure time of the sample are summed to obtain a sample sum value, and the average value of the sample sum value is used as the sample comprehensive speed.

[0040] In one embodiment of the present invention, v j Not equal to v j+1 .

[0041] In one embodiment of the present invention, the exposure time of one sample is equal to the exposure time of the vehicle-mounted lens in capturing the transmission image.

[0042] S222, divide the exposure time of the sample evenly into A r sample time slices, and based on v j The corresponding sample speed information is obtained v j Corresponding to A r List B of sample poses jr ={B jr，1 , B jr，2 ,…,B jr，g ,…,B jr，Ar}, thereby obtaining v j Corresponding sample pose list B j ={B j1 , B j2 ,…,B jr ,…,B js}, B jr，g It is v j Corresponding to A r The sample pose of the gth sample time slice, the value range of g is 1 to A r , positive integer A r is the number of time slices in the time slice number list, r ranges from 1 to s, and s is the number of time slices. Specifically, for each sample time slice, based on v j The sample rotation matrix is ​​obtained by integrating the sample angular velocity in the sample time slice, based on v j The sample displacement is obtained by quadratically integrating the sample linear acceleration at the sample time slice, and the sample pose is obtained based on the sample rotation matrix and the sample displacement.

[0043] S223, based on v j Corresponding sample pose list B j ={B j1 , B j2 ,…,B jr ,…,B js}, generate the sample point diffusion matrix list C j ={C j1, C j2 ,…,C jr ,…,C js}, and use C jr v j The corresponding sample transmission image is processed to obtain the sample clear image D jr , thereby obtaining v j Corresponding sample clear image list D j ={D j1 , D j2 ,…,D jr ,…,D js}, C jr It is v j The exposure time is divided into A r The sample point diffusion matrix obtained by the sample time slice.

[0044] Specifically, those skilled in the art know that any method of obtaining a point spread matrix based on posture in the prior art falls within the scope of protection of the present invention and will not be described in detail here.

[0045] Specifically, the sample clear image convolved with the sample point spread matrix plus the noise equals the sample clear image. Therefore, after obtaining the sample point spread matrix, the sample transmission image is deconvolved using the sample point spread matrix to obtain the sample clear image.

[0046] S224, get D jr The corresponding image quality value E jr Specifically, the image quality value is the peak signal-to-noise ratio value of the image, that is, the image quality value is the PSNR value of the image. PSNR is used to quantify the difference between the sample clear image and the sample transmitted image. The maximum value and mean square error value of the image pixels are used to measure the quality of the sample clear image, so that the sample clear image has better clarity at the processor level, although the PSNR value may not be consistent with human eye perception.

[0047] S225 , establishing a coordinate system with the sample integrated speed as the X-axis, the number of time slices as the Y-axis, and the image quality value as the Z-axis.

[0048] S226 , generating a three-dimensional image based on the sample comprehensive velocity list v, the time slice quantity list, and the image quality value.

[0049] Specifically, based on v j 、A r and E jr Constructing a three-dimensional image: Those skilled in the art know that any method of generating a three-dimensional image in the prior art falls within the protection scope of the present invention, such as using Python tools to generate a three-dimensional image.

[0050] S227 , obtaining the inflection point value of the three-dimensional image on the X axis as v0, and obtaining the inflection point value of the three-dimensional image on the Y axis as N0.

[0051] In summary, obtain the comprehensive speed list of samples and divide the exposure time of the samples evenly into A r sample time slices, and based on v j The corresponding sample speed information is obtained v j Corresponding to A r The sample pose list of D jr The corresponding image quality value E jr , with the sample comprehensive speed as the X-axis, the number of time slices as the Y-axis, and the image quality value as the Z-axis, a coordinate system is established. With the sample comprehensive speed as the X-axis, the number of time slices as the Y-axis, and the image quality value as the Z-axis, a coordinate system is established. The inflection point value of the three-dimensional image on the X-axis is obtained as v0, and the inflection point value of the three-dimensional image on the Y-axis is obtained as N0. The present invention uses different numbers of time slices for different sample comprehensive speeds to obtain the corresponding processed PSNR value, determines the speed threshold and the time slice number threshold through the inflection point, and obtains the speed threshold and the time slice number threshold more accurately.

[0052] Specifically, such as Figure 4 As shown, in one embodiment of the present invention, obtaining the inflection point value of the three-dimensional image on the X-axis as v0 in S227 also includes:

[0053] S2271, for A r , get the image quality value E jr With the sample integrated speed v j The first variation curve F r Those skilled in the art will appreciate that any method for obtaining a change curve based on coordinate values ​​in the prior art falls within the scope of protection of the present invention and will not be described in detail herein.

[0054] S2272, obtaining the first change curve F r The first inflection point AP r , get the X-axis coordinate value of AP1 to AP s The X-axis average coordinate value of the X-axis coordinate values ​​is used as v0.

[0055] Specifically, such as Figure 5 As shown, in one embodiment of the present invention, obtaining the inflection point value of the three-dimensional image on the Y axis as N0 in S227 also includes:

[0056] S2273, for v j , get the image quality value E jr Number of slices over time A r The second variation curve H j .

[0057] S2274, obtaining the second change curve H j The second turning point BP j , get the Y-axis coordinate value of BP1 to BP n The Y-axis average coordinate value of the Y-axis coordinate values ​​is taken as N0.

[0058] Specifically, such as Figure 6 As shown, in S300, based on the target vehicle-mounted lens pose corresponding to each target time slice, the target point spread matrix is ​​determined, which also includes:

[0059] S310: Project the target vehicle-mounted lens pose onto a 2D image plane to obtain projection points. Specifically, those skilled in the art will appreciate that any prior art method for projecting a pose onto a 2D image plane to obtain projection points falls within the scope of the present invention and will not be further described herein.

[0060] S320: Create an M×M blank point diffusion matrix, quantize the projected points to pixel coordinates based on the blank point diffusion matrix, and generate a target point diffusion matrix based on the pixel coordinates. Specifically, round the projected points to integer pixel locations, perform a quick check to ensure that the integer pixels are within the image, draw a trajectory so that the values ​​at the trajectory points accumulate to 1, and then normalize the PSF matrix to obtain the target point diffusion matrix.

[0061] Further, such as Figure 7 As shown, in S320, M is obtained through the following steps:

[0062] S321, obtain the first standard deviation α1 and the second standard deviation α2, α1 satisfies the following conditions: α1 = (f × (ra × T + 1 / 2 × a × T 2 )) / (z×s); α2 satisfies the following conditions: α2=s0×w×T, f is the focal length of the vehicle lens, ra is the vehicle speed, T is the exposure time, s0 is the maximum number of pixels from the center to the edge of the transmitted image, z is the preset object distance, w is the target angular velocity, s is the pixel resolution, and a is the target linear acceleration.

[0063] S322, obtain the composite standard deviation α0, α0 satisfies the following conditions: α0 = [k1 × (α1) 2 +k2×(α2) 2 ] 1 / 2 , k1 is the preset first weight, and k2 is the preset second weight.

[0064] In one embodiment of the present invention, k1=f×(T 2 / 2) / (z×s), k2=f×(d x 2 +d y2 ) 1 / 2 ×T / (z×s). d x is the size of the vehicle lens relative to the center of the vehicle in the preset X-axis direction, d y The size of the vehicle-mounted lens relative to the center of the vehicle in the preset Y-axis direction.

[0065] S323, obtain M, M satisfies the following condition: M=ceil(α0+k0×α3), ceil() is a rounding-up function, k0 is a preset third weight, and α3 is the size of a preset Gaussian kernel.

[0066] Specifically, the actual PSF matrix size needs to include a Gaussian kernel with a k0-fold expansion of the path points to avoid stage errors.

[0067] In one embodiment of the present invention, k0=3, and the preset Gaussian kernel is 1.5.

[0068] In summary, the first standard deviation α1 and the second standard deviation α2 are obtained, the composite standard deviation α0 is obtained, and M is obtained. The acquisition of M is more reasonable through the first standard deviation and the second standard deviation.

[0069] Further, such as Figure 8 As shown, in S200, after obtaining N, the following steps are also included:

[0070] S201, obtain a target prediction function, input cv and N into the target prediction function, and obtain a prediction result.

[0071] In one embodiment of the present invention, the target prediction function is γ / (1+e -f(cv,N) ), γ is a preset parameter. In one embodiment of the present invention, f(cv, N)=N+1 / (cv).

[0072] S202: If the prediction result does not meet the preset prediction requirement, N is updated based on the prediction result.

[0073] In one embodiment of the present invention, if γ / (1+e -f(cv,N) )<preset threshold, N is manually updated, and the threshold range of the preset threshold is 0 to 1.

[0074] An embodiment of the present invention also provides a non-transitory computer-readable storage medium, which can be set in an electronic device to store a computer program related to a method in the method embodiment. The computer program is loaded and executed by the processor to implement the method provided in the above embodiment.

[0075] An embodiment of the present invention further provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method provided in the above embodiment when executing the computer program.

[0076] Although some specific embodiments of the present invention have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It should also be understood by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention.

Claims

1. A method for correcting images transmitted by a vehicle-mounted lens, characterized in that: The method comprises the following steps: Acquire a transmission image captured by the vehicle-mounted lens, and acquire target velocity information of the vehicle-mounted lens during an exposure time for capturing the transmission image, wherein the target velocity information includes at least a target linear acceleration and a target angular velocity; Obtain N, and evenly divide the exposure time into N target time slices, and obtain the target vehicle-mounted lens pose corresponding to each target time slice based on the target speed information; Based on the target vehicle lens pose corresponding to each target time slice, the target point spread matrix is ​​determined, and the target point spread matrix is ​​used to process the transmitted image to obtain a clear target image; Among them, N is obtained through the following steps: Determine the target integrated velocity cv based on the target linear acceleration and the target angular velocity; If cv is greater than v0, obtain N, where N satisfies the following condition: N is equal to the average of all time slice numbers greater than N0 in the time slice number list, where the time slice number list includes several time slice numbers, v0 is the speed threshold, and N0 is the time slice number threshold; If cv is not greater than v0, obtain N, where N satisfies the following condition: N is equal to the average of all time slice numbers not greater than N0 in the time slice number list.

2. The method for correcting images transmitted by a vehicle-mounted lens according to claim 1, characterized in that: How to get v0 and N0: Get sample comprehensive speed list v = {v1, v2, ..., v j ,…,v n }, v j is the jth sample integrated speed, where j ranges from 1 to n, and n is the number of sample integrated speeds, the sample integrated speed being determined based on sample speed information of the sample vehicle during the exposure time of one sample, wherein the sample speed information includes at least sample linear acceleration and sample angular velocity; The exposure time of the sample is evenly divided into A r sample time slices, and based on v j The corresponding sample speed information is obtained v j Corresponding to A r List B of sample poses jr ={B jr，1 , B jr，2 ,…,B jr，g ,…,B jr，Ar }, thereby obtaining v j Corresponding sample pose list B j ={B j1 , B j2 ,…,B jr ,…,B js }, B jr，g It is v j Corresponding to A r The pose of the gth sample time slice, the value range of g is 1 to A r , positive integer A r is the number of the rth time slice in the time slice number list, r ranges from 1 to s, and s is the number of time slices; Based on v j Corresponding sample pose list B j ={B j1 , B j2 ,…,B jr ,…,B js }, generate the sample point diffusion matrix list C j ={C j1 , C j2 ,…,C jr ,…,C js }, and use C jr v j The corresponding sample transmission image is processed to obtain the sample clear image D jr , thereby obtaining v j Corresponding sample clear image list D j ={D j1 , D j2 ,…,D jr ,…,D js }, C jr It is v j The exposure time is divided into A r The sample point diffusion matrix obtained by the sample time slice; Get D jr The corresponding image quality value E jr ; A coordinate system is established with the sample integrated speed as the X-axis, the number of time slices as the Y-axis, and the image quality value as the Z-axis; Generate a three-dimensional image based on the sample integrated velocity list v, the time slice number list and the image quality value; The inflection point value of the three-dimensional image on the X axis is obtained as v0, and the inflection point value of the three-dimensional image on the Y axis is obtained as N0.

3. The method for correcting images transmitted by a vehicle-mounted lens according to claim 2, characterized in that: Get the inflection point value of the 3D image on the X axis as v0, and also include: To A r , get the image quality value E jr With the sample integrated speed v j The first variation curve F r ; Get the first change curve F r The first inflection point AP r , get the X-axis coordinate value of AP1 to AP s The X-axis average coordinate value of the X-axis coordinate values ​​is used as v0.

4. The method for correcting images transmitted by a vehicle-mounted lens according to claim 2, characterized in that: Obtaining the inflection point value of the 3D image on the Y axis as N0, further comprising: v j , get the image quality value E jr Number of slices over time A r The second variation curve H j ; Get the second change curve H j The second turning point BP j , get the Y-axis coordinate value of BP1 to BP n The Y-axis average coordinate value of the Y-axis coordinate values ​​is taken as N0.

5. The method for correcting images transmitted by a vehicle-mounted lens according to claim 1, characterized in that: Based on the target vehicle lens pose corresponding to each target time slice, the target point spread matrix is ​​determined, which also includes: Project the pose of each target vehicle-mounted camera onto the 2D image plane to obtain the projection point; Create a blank point diffusion matrix of size M×M, quantize the projection points to pixel coordinate points based on the blank point diffusion matrix, and generate a target point diffusion matrix based on the pixel coordinate points.

6. The method for correcting images transmitted by a vehicle-mounted lens according to claim 5, characterized in that: Obtain M by following the steps below: Obtain the first standard deviation α1 and the second standard deviation α2, where α1 satisfies the following conditions: α1 = (f × (ra × T + 1 / 2 × a × T 2 )) / (z×s); α2 satisfies the following conditions: α2=s0×w×T, where f is the focal length of the vehicle lens, ra is the vehicle speed, T is the exposure time, s0 is the maximum number of pixels from the center to the edge of the transmitted image, z is the preset object distance, w is the target angular velocity, s is the pixel resolution, and a is the target linear acceleration; Get the composite standard deviation α0, which satisfies the following conditions: α0 = [k1×(α1) 2 +k2×(α2) 2 ] 1 / 2 , k1 is the preset first weight, k2 is the preset second weight; Get M, where M satisfies the following condition: M=ceil(α0+k0×α3), where ceil() is a rounding-up function, k0 is a preset third weight, and α3 is the size of a preset Gaussian kernel.

7. The method for correcting images transmitted by a vehicle-mounted lens according to claim 1, characterized in that: After obtaining N, it also includes: Get the target prediction function, input cv and N into the target prediction function, and get the prediction result; If the prediction result does not meet the preset prediction requirements, N is updated based on the prediction result.

8. The method for correcting images transmitted by a vehicle-mounted lens according to claim 2, characterized in that: The image quality value is the peak signal-to-noise ratio value of the image.

9. A non-transitory computer-readable storage medium, characterized in that The storage medium stores a computer program, which is loaded and executed by a processor to implement the method for correcting images transmitted by a vehicle-mounted lens as described in any one of claims 1 to 8.

10. An electronic device comprising: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for correcting an image transmitted by a vehicle-mounted lens according to any one of claims 1 to 8 is implemented.

Images