Visual system compensation method and device in high-temperature environment and storage medium

Abstract

The invention relates to the field of chip production, and mainly relates to a visual system compensation method and device in a high-temperature environment and a storage medium, and the method comprises the following steps: S1, obtaining a visual system input image in the high-temperature environment; s2, detecting an image through an image processing algorithm; s3, the detected visual distortion is compensated; and S4, post-processing the compensated image, and outputting a final image. According to the invention, an image processing algorithm is adopted to detect visual distortion caused by a high-temperature effect in an image, a compensation model is introduced, and visual characteristic data of the model based on a high-temperature environment is trained, so that thermal deformation and optical distortion of an optical system, equipment offset and alignment errors caused by high temperature are compensated; the visual system is an effective method capable of realizing image quality compensation, optical deformation correction and alignment error in a high-temperature environment.

Description

Technical Field

[0001] The present invention relates to the field of chip production, and mainly relates to a method, a device and a storage medium for visual system compensation in a high temperature environment. Background Art

[0002] With the continuous advancement of semiconductor technology, the demand for image processing technology in the chip manufacturing process is increasing, especially in the precision inspection and quality control of chip production, the visual system has become an indispensable tool. The visual inspection system can monitor and analyze various defects in the chip production process in real time, such as wafer surface defects, mask alignment errors, inter-layer alignment problems, etc., to ensure the accuracy and yield of the product.

[0003] However, many links in the chip manufacturing process, especially the manufacturing process in high temperature environment, pose great challenges to the vision system. Chip manufacturing usually involves high temperature heat treatment processes (such as annealing, welding, chemical vapor deposition, bonding, etc.), which will cause production equipment, sensors and optical components to operate in a high temperature environment, leading to a series of problems:

[0004] 1. Thermal deformation and optical distortion of the optical system: High temperature may cause thermal expansion of lenses, optical elements and other key components, resulting in distortion of the optical system or focal length change. Especially in micron-level chip manufacturing, even tiny imaging distortion may lead to failure or misjudgment of defect detection.

[0005] 2. Equipment offset and alignment errors caused by temperature: In high temperature environments, the mechanical components of chip production equipment may shift or deform due to thermal expansion. This offset can cause alignment errors in the visual system, making the chip features in the image inconsistent with the actual position, thereby affecting the accuracy of the automated inspection system.

[0006] Most existing high temperature vision compensation technologies rely on hardware improvements, such as using high temperature stable optical materials, improving sensor design and heat dissipation systems, or reducing ambient temperature by adding cooling systems. However, these methods still cannot completely solve the effects of image distortion, thermal deformation, and temperature non-uniformity at high temperatures.

[0007] Therefore, how to deal with the impact of high temperature environment on the visual system during chip production has become a technical problem that needs to be solved urgently. An effective method that can achieve image quality compensation, correct optical deformation and alignment errors in a high temperature environment can not only improve the yield and accuracy of chip manufacturing, but also reduce the scrap rate in the production process, and improve production efficiency and cost-effectiveness. Summary of the invention

[0008] In order to solve the deficiencies in the prior art, the present invention proposes a visual system compensation method in a high temperature environment, which can effectively and timely deal with problems such as image degradation, optical distortion and equipment errors caused by high temperature, and ensure that high-precision detection capabilities can be maintained in a high-temperature production environment, thereby greatly improving the level of automated quality control in the chip production process. The specific technical solution is as follows:

[0009] A method for compensating a visual system in a high temperature environment, characterized by comprising the following steps:

[0010] S1: Obtaining the visual system input image in a high temperature environment;

[0011] S2: Detect visual distortion caused by high temperature effects in images through image processing algorithms;

[0012] S3: using a pre-trained compensation model to compensate for the detected visual distortion, wherein the compensation model is trained based on visual characteristic data in a high temperature environment;

[0013] S4: Post-process the compensated image and output the final image.

[0014] Preferably, the image processing algorithm includes a combination of any one or more of ray tracing and refraction simulation, edge detection and deformation analysis, image registration algorithm, geometric transformation detection method, frequency domain analysis method and image quality assessment method to detect image distortion caused by high temperature effects.

[0015] Preferably, the visual distortion includes image color deviation, image distortion and image blur.

[0016] Preferably, in S3, the training method of the compensation model comprises the following steps:

[0017] S3.1. Collect image data under different high temperature environments, including the correlation between temperature and visual distortion;

[0018] S3.2. Select deep learning methods to simulate and compensate for visual distortion;

[0019] S3.3, using the collected high temperature data to train the compensation model, so that the compensation model can learn to identify and correct visual distortion, and obtain a trained compensation model;

[0020] S3.4. Apply the trained compensation model to real-time image data to automatically correct image distortion.

[0021] Preferably, in S4, the compensated image is post-processed, and the specific processing method includes any one or a combination of multiple methods of noise removal, contrast adjustment, color correction, sharpening, cropping and edge repair, and format conversion.

[0022] Preferably, the present invention further comprises the following steps:

[0023] S5. Based on the temperature distribution of different regions in the input image, dynamically adjust the compensation model parameters to adapt to the visual distortion under different temperature conditions.

[0024] The present invention also provides a device for applying the visual system compensation method in the above-mentioned high temperature environment, comprising:

[0025] An input module, used to obtain an input image of a visual system in a high temperature environment;

[0026] A processing module for executing image processing algorithms to detect visual distortion caused by high temperature effects;

[0027] A compensation module, used to compensate the detected visual distortion through a pre-trained compensation model;

[0028] The output module is used to output the final image after compensation.

[0029] Preferably, the device also includes a temperature sensor for real-time monitoring of ambient temperature and dynamic adjustment of compensation model parameters.

[0030] Preferably, the processing module includes an image filter, an edge detection module and a contrast enhancement module to improve the distortion detection accuracy of the image.

[0031] The present invention also provides a computer storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned visual system compensation method in a high temperature environment.

[0032] Compared with the prior art, the visual system compensation method in a high temperature environment described in the present invention has the following characteristics:

[0033] 1. The visual system compensation method in a high temperature environment described in the present invention adopts an image processing algorithm to detect the visual distortion caused by the high temperature effect in the image, and refers to a compensation model and trains the model based on the visual characteristic data in a high temperature environment to compensate for the thermal deformation and optical distortion of the optical system caused by high temperature, as well as the device offset and alignment error. This method enables the visual system to achieve image quality compensation and correct optical deformation and alignment error in a high temperature environment. This method can not only improve the yield and accuracy of chip manufacturing, but also reduce the scrap rate in the production process, and improve production efficiency and cost-effectiveness.

[0034] 2. The visual system compensation method in a high temperature environment described in the present invention specifically combines any one or more methods including ray tracing and refraction simulation, edge detection and deformation analysis, image registration algorithm, geometric transformation detection method, frequency domain analysis method and image quality assessment method to better detect image distortion caused by high temperature effects.

[0035] 3. The visual system compensation method in a high temperature environment described in the present invention also performs post-processing after compensating the image. The specific processing method includes a combination of any one or more methods of noise removal, contrast adjustment, color correction, sharpening, cropping and edge repair, and format conversion to improve the accuracy of the image. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0037] Figure 1 is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0040] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, the singular forms "a", "an" and "the" are intended to include plural forms unless the context clearly indicates otherwise.

[0041] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0042] See also Figure 1 , a visual system compensation method in a high temperature environment, characterized in that it includes the following steps:

[0043] S1: Obtaining the visual system input image in a high temperature environment;

[0044] S2: Detect visual distortion caused by high temperature effects in images through image processing algorithms;

[0045] S3: using a pre-trained compensation model to compensate for the detected visual distortion, wherein the compensation model is trained based on visual characteristic data in a high temperature environment;

[0046] S4: Post-process the compensated image and output the final image.

[0047] Among them, the visual system includes an acquisition system and a computer system. The acquisition system can be a high-speed camera, an intelligent monitor, etc. The image collected by the visual system enters the computer system, and the computer system detects the visual distortion caused by the high temperature effect in the image through the image processing algorithm. Specifically, the visual distortion includes image color deviation, image distortion and image blur; then the detected visual distortion is compensated by a pre-trained compensation model. The compensation model is trained based on the visual characteristic data in a high temperature environment to obtain a compensated image, and then the compensated image is post-processed to output the final image.

[0048] Furthermore, the image processing algorithm includes the following specific schemes:

[0049] (1) Thermal effect and refraction model;

[0050] High temperature effects are often caused by changes in the refractive index due to differences in air temperature, especially near heat sources, where temperature gradients can cause light to refract. Physical model-based detection methods can identify distortions by simulating and analyzing refraction effects in images. Refraction effects can cause edges of objects in images to be distorted, ghosted, or blurred. To detect these distortions, the following methods can be used:

[0051] (1.1) Ray tracing and refraction simulation: Use the light refraction caused by the hot air layer to simulate the distortion of objects in the image, thereby predicting and identifying visual distortion caused by temperature gradient changes.

[0052] (1.2) Edge detection and deformation analysis: Distortion in an image will affect the edge and shape of the object. Edge detection algorithms (such as Canny, Sobel, etc.) can be used to extract the edges in an image and analyze whether there are irregular distortions or deformations.

[0053] (2) Detection methods based on image distortion;

[0054] Visual distortion caused by high temperature effects can lead to changes in the shape or position of objects in the image. Common methods for detecting image distortion include:

[0055] (2.1) Image registration: Align the original image with the ideal image or reference image through an image registration algorithm (such as feature point-based registration method). If the image is distorted due to thermal effects, the registration process will reveal the difference in shape.

[0056] (2.2) Geometric distortion detection: The image is analyzed using geometric transformation detection methods (such as affine transformation, perspective transformation, etc.). By estimating the geometric shape changes of objects in the image, distortion caused by temperature effects can be detected.

[0057] (3) Frequency domain analysis;

[0058] Visual distortion caused by high temperature effects often has certain frequency characteristics. Frequency domain analysis can help detect the characteristics of these distortions. Common frequency domain methods include:

[0059] (3.1) Fourier transform: By performing Fourier transform on an image, the image can be converted from the spatial domain to the frequency domain. In the frequency domain, the distortion caused by thermal effects appears as high-frequency or low-frequency noise. By analyzing the spectrum, it is possible to detect whether there is distortion caused by thermal effects in the image.

[0060] (3.2) Wavelet transform: Wavelet transform can be used for multi-scale analysis. By analyzing the frequency components at different scales, it can effectively identify the local distortion or deformation caused by high temperature effects in the image.

[0061] (4) Image quality assessment

[0062] Since high temperature effects may cause blurring or loss of details in images, some common image quality assessment methods can be used to quantify the visual distortion of images:

[0063] (4.1) Structural Similarity Index (SSIM): SSIM is used to measure the structural similarity of images and is suitable for detecting structural distortion of images caused by high temperature effects. If the SSIM value between the original image and the image affected by high temperature drops significantly, it means that the image has been distorted.

[0064] (4.2) Peak Signal-to-Noise Ratio (PSNR): PSNR is used to measure the quality of an image. A lower PSNR value indicates a poorer image quality. If the image is distorted due to high temperature effects, the PSNR value will be lower, which may appear as a blurred image or reduced contrast.

[0065] (5) Deep Learning Methods

[0066] With the widespread application of deep learning, methods based on convolutional neural networks (CNN) can automatically learn the distortion features in images. A convolutional neural network can be trained to identify distortion caused by high temperature effects. The steps of the deep learning method include:

[0067] (5.1) Dataset collection and annotation: Collect image datasets containing visual distortion caused by thermal effects and annotate the distorted areas.

[0068] (5.2) Network training and recognition: Use a deep neural network (such as CNN or U-Net) to train the model and identify the distortion caused by thermal effects in the image. The advantage of this method is that it can automatically identify complex thermal effect features without the need to manually design a specific image processing algorithm.

[0069] (6) Image restoration based on fluid dynamics simulation

[0070] In some cases, visual distortion caused by thermal effects can be restored by simulating fluid mechanics and temperature gradients. For example, a fluid simulation model based on the Navier-Stokes equations can be used to simulate the effect of temperature changes in the air on light propagation, thereby correcting the image.

[0071] (7) Local area analysis

[0072] Since the distortion caused by high temperature effects is limited to local areas in the image (such as near the heat source or where the temperature difference is large), the impact of thermal effects can be identified by analyzing the local areas of the image. For example:

[0073] (7.1) Local contrast analysis: By analyzing the contrast changes in local areas of the image, visual distortion caused by high temperature effects can be detected.

[0074] (7.2) Local geometric feature analysis: By analyzing the changes in local shape features (such as circles, straight lines, etc.), analyze whether there is any geometric deformation caused by thermal effects.

[0075] According to the actual situation, this embodiment can select one or more solutions from the above solutions for combination, which can effectively detect and analyze the visual distortion caused by high temperature effects. Image registration, frequency domain analysis, deep learning and other methods can be used to identify the specific type and location of distortion. For specific application scenarios, it may be necessary to combine physical models with image processing technology for joint analysis to obtain more accurate detection results.

[0076] Furthermore, the training method of the compensation model includes the following steps:

[0077] S3.1. Collect image data under different high temperature environments, including the correlation between temperature and visual distortion;

[0078] S3.2. Select deep learning methods to simulate and compensate for visual distortion;

[0079] S3.3, using the collected high temperature data to train the compensation model, so that the compensation model can learn to identify and correct visual distortion, and obtain a trained compensation model;

[0080] S3.4. Apply the trained compensation model to real-time image data to automatically correct image distortion.

[0081] The deep learning methods mentioned above include convolutional neural networks (CNN) or generative adversarial networks (GAN). Both convolutional neural networks and generative adversarial networks are very powerful models in the field of image processing and can be used to handle training tasks for visually distorted images. These two network architectures can be used to repair or enhance the quality of distorted images (for example, image dedistortion, image super-resolution, image restoration, etc.). The following describes how to use CNN and GAN to train visually distorted images:

[0082] Convolutional neural networks (CNNs) are used for three tasks: denoising, dedistortion, and image restoration. Denoising is to restore a clean image from a noisy image; dedistortion is to repair image distortion caused by optical distortion (such as lens distortion) or geometric transformation; and image restoration is to repair missing or damaged image areas.

[0083] Convolutional neural network (CNN) consists of multiple convolutional layers and pooling layers, which can extract useful features from the input image. When dealing with visual distortion, the goal of the network is to learn how to restore the original clear image through these features. For dedistortion tasks, the structure of the convolutional neural network can use the U-Net architecture, which is very suitable for image restoration and denoising because it has an encoder-decoder structure that can capture the global and local features of the image.

[0084] Convolutional neural networks (CNNs) use loss functions: Common loss functions include mean square error (MSE), which aims to minimize the difference between the generated image and the target image. For more complex tasks (such as image denoising), perceptual loss can also be used to improve visual quality.

[0085] The optimizer for the convolutional neural network CNN uses common optimization algorithms, such as the Adam optimizer, to adjust the network weights.

[0086] The training process is as follows:

[0087] The distorted image is input, and the network extracts features through the convolution layer and generates a predicted clear image. Then, the network parameters are updated by calculating the loss function until the generated image is close enough to the real clear image.

[0088] Generative adversarial networks (GANs) consist of a generator and a discriminator, and use adversarial training to improve the quality of image generation. Generative adversarial networks (GANs) are particularly suitable for generating realistic images and can be used to solve tasks such as image dedistortion and image restoration. The task of the generator is to generate a result that is as close to a real and clear image as possible from a distorted image. It is composed of a convolutional neural network (CNN), with a distorted image as input and a restored image as output; the task of the discriminator is to determine whether the image is a real image (from a data set) or a fake image generated by the generator. It is also a convolutional neural network (CNN), which outputs a probability value indicating the probability that the input image is a real image. Through adversarial training, the generator will continue to improve to deceive the discriminator and make the generated images more and more realistic.

[0089] The discriminator and generator of the Generative Adversarial Network (GAN) compete with each other. The generator hopes to "cheat" the discriminator to generate real images, while the discriminator tries to distinguish between real and fake images. On this basis, additional reconstruction losses are used to help the generator restore image details more accurately. For example, L1 loss can help the generated image to be as close to the real image as possible. In order to improve the visual effect of the image, a perceptual loss function can be introduced to optimize the structure and texture of the image by comparing high-level convolution features (rather than pixel-level errors).

[0090] For the training of the generator, the generator can continuously improve its output by minimizing the difference between the generated image and the real image; while for the training of the discriminator, it is optimized by maximizing its ability to distinguish between real and fake images.

[0091] Generative adversarial networks (GANs) can generate more realistic and delicate images, which are particularly suitable for image restoration tasks. Through adversarial training, they can effectively reduce the blurring problems that may be encountered in traditional convolutional neural networks (CNNs), and the generated images are of higher quality.

[0092] Furthermore, in S4, the compensated image is post-processed, and the specific processing method mainly includes the following aspects:

[0093] (1) Noise removal: Noise that may be introduced during the compensation process is removed through filtering (such as Gaussian filtering, median filtering, etc.).

[0094] (2) Contrast adjustment: Adjust the contrast of the image as needed to improve the visibility of details.

[0095] (3) Color correction: If the compensation process involves color changes, color mapping or white balance correction can be used to make the image color more realistic.

[0096] (4) Sharpening: Enhance image details through sharpening algorithms (such as Laplacian filtering and Unsharp Mask).

[0097] (5) Cropping and edge repair: If compensation causes image edge distortion or black edges, you can crop the image or perform edge repair.

[0098] (6) Format conversion: Save the image to the required file format, such as JPEG, PNG, etc.

[0099] For the case where the temperature of each part of the environment is uneven or unstable, the present invention further includes the following steps for dynamically adjusting the compensation model parameters:

[0100] S5. Based on the temperature distribution of different regions in the input image, dynamically adjust the compensation model parameters to adapt to the visual distortion under different temperature conditions.

[0101] The present invention also provides a device for applying the visual system compensation method in the above-mentioned high temperature environment, comprising:

[0102] An input module, used to obtain an input image of a visual system in a high temperature environment;

[0103] A processing module for executing image processing algorithms to detect visual distortion caused by high temperature effects;

[0104] A compensation module, used to compensate the detected visual distortion through a pre-trained compensation model;

[0105] The output module is used to output the final image after compensation.

[0106] The device can be used to apply the above-mentioned visual system compensation method in a high temperature environment, which can not only improve the yield and accuracy of chip manufacturing, but also reduce the scrap rate in the production process and improve production efficiency and cost-effectiveness.

[0107] Furthermore, in case of uneven and unstable temperatures in various parts of the environment, the device also includes a temperature sensor for real-time monitoring of the ambient temperature and dynamically adjusting the compensation model parameters.

[0108] Furthermore, the processing module includes an image filter, an edge detection module and a contrast enhancement module to improve the distortion detection accuracy of the image.

[0109] The present invention also provides a computer storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned visual system compensation method in a high temperature environment.

[0110] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0111] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0112] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0113] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for compensating a visual system in a high temperature environment, characterized in that: The following steps are involved: S1: Obtaining the visual system input image in a high temperature environment; S2: Detect visual distortion caused by high temperature effects in images through image processing algorithms; S3: using a pre-trained compensation model to compensate for the detected visual distortion, wherein the compensation model is trained based on visual characteristic data in a high temperature environment; S4: Post-process the compensated image and output the final image.

2. The method for compensating a visual system in a high temperature environment according to claim 1, characterized in that: The image processing algorithm includes any one or more combinations of ray tracing and refraction simulation, edge detection and deformation analysis, image registration algorithms, geometric transformation detection methods, frequency domain analysis methods, and image quality assessment methods to detect image distortion caused by high temperature effects.

3. The method for compensating a visual system in a high temperature environment according to claim 1 or 2, characterized in that: Visual distortion includes image color deviation, image distortion and image blur.

4. The method for compensating a visual system in a high temperature environment according to claim 1, characterized in that: In S3, the training method of the compensation model comprises the following steps: S3.

1. Collect image data under different high temperature environments, including the correlation between temperature and visual distortion; S3.

2. Select deep learning methods to simulate and compensate for visual distortion; S3.3, using the collected high temperature data to train the compensation model, so that the compensation model can learn to identify and correct visual distortion, and obtain a trained compensation model; S3.

4. Apply the trained compensation model to real-time image data to automatically correct image distortion.

5. The method for compensating a visual system in a high temperature environment according to claim 1, characterized in that: In S4, the compensated image is post-processed, and the specific processing method includes any one or a combination of multiple methods of noise removal, contrast adjustment, color correction, sharpening, cropping and edge repair, and format conversion.

6. The method for compensating a visual system in a high temperature environment according to claim 1, characterized in that: Further comprising the steps of: S5. Based on the temperature distribution of different regions in the input image, dynamically adjust the compensation model parameters to adapt to the visual distortion under different temperature conditions.

7. A visual system compensation device in a high temperature environment, characterized in that: The method for compensating a visual system in a high temperature environment as claimed in any one of claims 1 to 6 comprises: An input module, used to obtain an input image of a visual system in a high temperature environment; A processing module for executing image processing algorithms to detect visual distortion caused by high temperature effects; A compensation module, used to compensate the detected visual distortion through a pre-trained compensation model; The output module is used to output the final image after compensation.

8. The visual system compensation device in a high temperature environment according to claim 7, characterized in that: It also includes a temperature sensor for real-time monitoring of ambient temperature and dynamic adjustment of compensation model parameters.

9. The visual system compensation device in a high temperature environment according to claim 7, characterized in that: The processing module includes an image filter, an edge detection module and a contrast enhancement module to improve the distortion detection accuracy of the image.

10. A computer storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the visual system compensation method in a high temperature environment according to any one of claims 1 to 6.

Images