A material optimization method and device based on resin lens collision detection

Abstract

The present invention discloses a material optimization method and device based on resin lens collision detection, which relates to the field of artificial intelligence technology. The method includes: extracting elements from the lens collision application scenario information to obtain collision scenario application parameter information; performing data detection on the collision scenario application parameter information to obtain the resin lens collision detection result; obtaining the lens material optimization requirement information according to the resin lens collision detection result and the resin lens material parameter information; constructing a material optimization fitness function based on the lens material optimization requirement information; obtaining the lens material optimization search space according to the resin lens material parameter information; performing optimization within the lens material optimization search space based on the material optimization fitness function, outputting a material optimization guidance plan, and performing lens material optimization management according to the material optimization guidance plan. The technical effects of realizing the autonomous optimization of collision detection, improving the accuracy of lens material optimization, and ensuring the application performance of resin lenses are achieved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and particularly to a method and device for optimizing materials based on collision detection of resin lenses. Background Art

[0002] A resin lens is a lens made of organic materials, which is chemically synthesized from resin as the raw material and processed and polished. Its internal structure is a three-dimensional network structure formed by the connection of polymer chain structures. The intermolecular structure of resin lenses is relatively loose, with the advantages of low density, good light transmittance, and strong impact resistance. Therefore, to ensure the performance of the lenses, it is of great application significance to optimize the materials of resin lenses.

[0003] However, the existing technology has low accuracy in optimizing the material components, resulting in technical problems that affect the application performance of resin lenses. Summary of the Invention

[0004] This application provides a method and device for optimizing materials based on collision detection of resin lenses, which solves the technical problem that the existing technology has low accuracy in optimizing the material components, resulting in affecting the application performance of resin lenses. By conducting scenario collision tests on resin lenses, determining the material optimization parameter scheme, realizing the autonomous optimization of collision detection, improving the accuracy of lens material optimization, and further ensuring the application performance of resin lenses.

[0005] In view of the above problems, the present invention provides a method and device for optimizing materials based on collision detection of resin lenses.

[0006] In a first aspect, this application provides a method for optimizing materials based on collision detection of resin lenses. The method includes: obtaining information on the lens collision application scenario; extracting elements from the information on the lens collision application scenario to obtain collision scenario application parameter information; performing data detection through the collision scenario application parameter information to obtain the resin lens collision detection result; obtaining resin lens material parameter information, and based on the resin lens collision detection result and the resin lens material parameter information, obtaining lens material optimization requirement information; constructing a material optimization fitness function based on the lens material optimization requirement information; obtaining the lens material optimization search space according to the resin lens material parameter information; performing optimization within the lens material optimization search space based on the material optimization fitness function, outputting a material optimization guidance scheme, and performing lens material optimization management according to the material optimization guidance scheme.

[0007] On the other hand, the present application also provides a material optimization device based on resin lens collision detection, and the device includes: a scenario acquisition module for acquiring lens collision application scenario information; a feature extraction module for extracting features from the lens collision application scenario information to obtain collision scenario application parameter information; a data detection module for performing data detection through the collision scenario application parameter information to obtain a resin lens collision detection result; an optimization requirement acquisition module for obtaining resin lens material parameter information, and acquiring lens material optimization requirement information according to the resin lens collision detection result and the resin lens material parameter information; a fitness construction module for constructing a material optimization fitness function based on the lens material optimization requirement information; an optimization space acquisition module for obtaining a lens material optimization space according to the resin lens material parameter information; an optimization management module for performing optimization within the lens material optimization space based on the material optimization fitness function, outputting a material optimization guidance plan, and performing lens material optimization management according to the material optimization guidance plan.

[0008] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0009] By adopting the technical solution of extracting features from the lens collision application scenario information, then performing data detection through the extracted collision scenario application parameter information to obtain a resin lens collision detection result, acquiring lens material optimization requirement information according to the resin lens collision detection result and the resin lens material parameter information, then constructing a material optimization fitness function based on the lens material optimization requirement information, obtaining a lens material optimization space according to the resin lens material parameter information, performing optimization within the lens material optimization space based on the material optimization fitness function, outputting a material optimization guidance plan, and performing lens material optimization management according to the material optimization guidance plan. Furthermore, the technical effect of determining a material optimization parameter solution through a resin lens for scenario collision testing, realizing the autonomous optimization of collision detection, improving the accuracy of lens material optimization, and thus ensuring the application performance of resin lenses is achieved. Description of the Drawings

[0010] Figure 1 It is a flowchart of a material optimization method based on resin lens collision detection according to the present application;

[0011] Figure 2 It is a flowchart of obtaining a resin lens collision detection result in a material optimization method based on resin lens collision detection according to the present application;

[0012] Figure 3 It is a flowchart of obtaining lens collision detection defect information in a material optimization method based on resin lens collision detection according to the present application;

[0013] Figure 4 This is a schematic structural diagram of a material optimization device for collision detection of resin lenses in the present application;

[0014] Explanation of reference numerals: Scene acquisition module 11, element extraction module 12, data detection module 13, optimization requirement acquisition module 14, fitness construction module 15, optimization space acquisition module 16, optimization management module 17. Detailed implementation manners

[0015] The present application provides a material optimization method and device based on collision detection of resin lenses, solves the technical problem of low accuracy in optimizing the material components, which affects the application performance of resin lenses, and achieves the technical effect of performing scene collision tests through resin lenses, determining the material optimization parameter scheme, realizing the autonomous optimization of collision detection, improving the accuracy of lens material optimization, and further ensuring the application performance of resin lenses.

[0016] Embodiment 1

[0017] As Figure 1 shown, the present application provides a material optimization method based on collision detection of resin lenses, and the method includes:

[0018] Step S100: Obtain the information of the lens collision application scenario;

[0019] Step S200: Extract elements from the information of the lens collision application scenario to obtain the collision scenario application parameter information;

[0020] Specifically, a resin lens is a lens made of organic materials, which is chemically synthesized from resin as the raw material and processed and polished. Its internal structure is a three-dimensional network structure formed by the connection of polymer chain structures. The intermolecular structure of resin lenses is relatively loose, with the advantages of low density, good light transmittance, and strong impact resistance for optical resin lenses. Therefore, to ensure the use performance of the lenses, it is of important application significance to optimize the materials of resin lenses.

[0021] To simulate the resin lens collision test and obtain the information of the lens collision application scenario, for example, the lens falls from a high altitude, the lens is hit by a ball, etc. Then, extract elements from the information of the lens collision application scenario, that is, extract the collision parameters of the lens to obtain the collision scenario application parameter information, including impact speed, collision intensity, collision angle, etc., to ensure the accuracy and comprehensiveness of the lens collision test.

[0022] Step S300: Perform data detection through the collision scenario application parameter information to obtain the resin lens collision detection result;

[0023] As Figure 2As shown, further, the data detection is performed by using the parameter information of the collision scenario to obtain the collision detection result of the resin lens. Step S300 of this application further includes:

[0024] Step S310: Perform orthogonal arrangement on the parameter information of the collision scenario to obtain a collision test scenario parameter table;

[0025] Step S320: Perform a lens collision test based on the parameter information of the collision scenario to obtain lens collision detection data information and lens collision picture acquisition information;

[0026] Step S330: Extract feature points from the lens collision picture acquisition information to obtain lens collision detection defect information;

[0027] Step S340: Based on the lens collision detection data information and the lens collision detection defect information, obtain the collision detection result of the resin lens.

[0028] As Figure 3 shown, further, the step of extracting feature points from the lens collision picture acquisition information to obtain lens collision detection defect information, step S330 of this application further includes:

[0029] Step S331: Build a lens collision defect detection model, which includes an input layer, an image recognition layer, a defect classification layer, and an output layer;

[0030] Step S332: Take the lens collision picture acquisition information as the input layer and input it into the image recognition layer to obtain lens collision feature information;

[0031] Step S333: Input the lens collision feature information into the defect classification layer to obtain the lens collision detection defect information;

[0032] Step S334: Output the lens collision detection defect information as the output result through the output layer.

[0033] Further, the step of taking the lens collision picture acquisition information as the input layer and inputting it into the image recognition layer to obtain lens collision feature information, step S332 of this application further includes:

[0034] Step S3321: Obtain predetermined image convolution features according to the lens smoothness standard;

[0035] Step S3322: Input the lens collision picture acquisition information into the image recognition layer for feature extraction;

[0036] Step S3323: Obtain the output information of the image recognition layer, where the output information is the lens collision feature information that does not conform to the predetermined image convolution feature.

[0037] Specifically, data detection is performed through the collision scenario application parameter information. Specifically, orthogonal arrangement is performed on the collision scenario application parameter information, that is, each collision scenario application parameter is arranged and combined to form different test parameter schemes, namely the collision test scenario parameter table. Based on the collision scenario application parameter information, a lens collision test is performed, and the test situation is monitored and collected, including lens collision detection data information, that is, surface, optical, and geometric test data quantitatively measured by measuring instruments such as calipers, vertex height gauges, thickness gauges, and curvature meters; and lens collision picture acquisition information, which is lens collision picture information obtained by image acquisition devices such as cameras or industrial cameras, including information such as the lens surface, structure, and color.

[0038] Feature points are extracted from the lens collision picture acquisition information. Specifically, a lens collision defect detection model is built. The lens collision defect detection model is used to extract defect points from the lens collision image and includes an input layer, an image recognition layer, a defect classification layer, and an output layer. The lens collision picture acquisition information is used as the input layer and input into the image recognition layer. The image recognition layer is used to recognize the surface defect features of the lens. First, according to the lens smoothness standard, the predetermined image convolution feature, that is, the convolution feature value of the lens smoothness standard, is determined. The lens collision picture acquisition information is input into the image recognition layer for feature extraction, that is, convolution operation is performed on the image features to obtain the output information of the image recognition layer. The output information is the lens collision feature information that does not conform to the predetermined image convolution feature, that is, the defect flaw surface feature that does not conform to the surface smoothness standard.

[0039] The lens collision feature information is input into the defect classification layer. The defect classification layer is used to classify the recognized collision defect features and can be compared with the lens defect database to obtain the lens collision detection defect information, that is, the specific defect categories of the lens defects, such as impurities, scratches, bubbles, etc. The lens collision detection defect information is output as the output result through the output layer. Based on the lens collision detection data information and the lens collision detection defect information, the resin lens collision detection result is jointly determined. By combining the lens data test result and the collision picture feature analysis result to determine the lens collision detection result, the resin lens collision detection result is made more reasonable and accurate, thereby improving the accuracy of lens material optimization.

[0040] Step S400: Obtain the resin lens material parameter information. According to the resin lens collision detection result and the resin lens material parameter information, obtain the lens material optimization requirement information;

[0041] Specifically, resin lens material parameter information is obtained by analyzing the lens material. The resin lens material parameter information is the composition of the resin lens material for the impact test, including parameters such as material composition, material type, and material content. Based on the resin lens impact detection result and the resin lens material parameter information, lens material optimization requirement information is determined, that is, the target of material optimization requirement to be carried out based on the resin lens material composition.

[0042] Step S500: Based on the lens material optimization requirement information, construct a material optimization fitness function;

[0043] Step S600: Based on the resin lens material parameter information, obtain the lens material optimization space;

[0044] Specifically, based on the lens material optimization requirement information, a material optimization fitness function is constructed. The optimization fitness function is the main index describing the material optimization performance. Through the optimization fitness, the target material is evaluated and optimized, and optimization can be carried out by using algorithms such as simulated annealing algorithm and genetic algorithm. Based on the resin lens material parameter information, the lens material optimization space is obtained. The lens material optimization space is the set of all possible parameters for optimizing the material parameters under the test lens material parameter information.

[0045] Step S700: Optimize within the lens material optimization space based on the material optimization fitness function, output a material optimization guidance plan, and perform lens material optimization management according to the material optimization guidance plan.

[0046] Furthermore, for the optimization within the lens material optimization space based on the material optimization fitness function to output a material optimization guidance plan, step S700 of this application further includes:

[0047] Step S710: Randomly select material optimization parameters from the lens material optimization space as the optimal material parameters;

[0048] Step S720: Calculate the optimization fitness of the optimal material parameters according to the material optimization fitness function;

[0049] Step S730: Randomly select material optimization parameters from the lens material optimization space as the comparison parameters;

[0050] Step S740: Calculate the comparison fitness of the comparison parameters according to the material optimization fitness function;

[0051] Step S750: If the comparison fitness is greater than the optimization fitness, then replace the comparison parameters as the optimal material parameters;

[0052] Step S760: If the optimal parameters of the material do not change during the iterative optimization for the threshold number of times or the iterative optimization reaches the preset number of times, then output the optimal parameters of the material to obtain the optimized guidance scheme for the said material.

[0053] Furthermore, step S760 of the present application further includes:

[0054] Step S761: If the comparison fitness is less than the optimization fitness, then replace the comparison parameters with the optimal parameters of the material according to a probability, and the probability is calculated by the following formula:

[0055]

[0056] where r2 is the comparison fitness, r1 is the optimization fitness, and k is the optimization speed factor.

[0057] Specifically, optimization is performed based on the material optimization fitness function within the optimization search space of the lens material. Specifically, first randomly select a material optimization parameter from the optimization search space of the lens material and use it as the optimal parameter of the material. According to the material optimization fitness function, calculate the optimization fitness of the optimal parameter of the material. The higher the fitness, the better the optimization performance of the lens material. Then randomly select another material optimization parameter from the optimization search space of the lens material as the comparison parameter, and calculate the comparison fitness of the comparison parameter according to the material optimization fitness function. Compare the comparison fitness and the optimization fitness. If the comparison fitness is greater than the optimization fitness, then replace the comparison parameter with the optimal parameter of the material.

[0058] If the comparison fitness is less than the optimization fitness, then it is necessary to judge the acceptance probability. Calculate the acceptance probability through the probability formula where r2 is the comparison fitness, r1 is the optimization fitness, and k is the optimization speed factor. It can be seen from this that the acceptance probability is related to the difference between the comparison fitness and the optimization fitness. k is a constant that gradually decreases with the number of optimization iterations. In the initial stage of the optimization search, k is larger, and the optimal parameter of the material is probably not the global optimal parameter of the material and may be a local optimum. To avoid the optimization process stagnating at the local optimum, k is larger so that P is larger, and with a greater probability, accept the relatively inferior comparison parameter as the optimal solution. In the later stage of the optimization search, the current optimal parameter of the material is probably the global optimal parameter of the material. To improve the accuracy of the optimization search, k is smaller so that P is smaller, and with a smaller probability, accept the relatively inferior optimization parameter as the global optimal parameter of the material to improve the accuracy of the optimization search. Optionally, the decreasing method of k can be exponential decrease or logarithmic decrease or any other existing decreasing methods, and the value of k and the decreasing method can be determined according to the number of optimization parameters.

[0059] Further, a threshold number of times is set, and the threshold number of times is the iteration number limit. If, after repeated iterations until the threshold number of times is satisfied, the optimal material parameters no longer change, then they are output as the optimal material parameters. Or, when the number of iterative optimizations reaches the preset number, the optimal material parameters are output, thereby determining the material optimization guidance scheme, that is, the lens material optimization parameter scheme. It achieves the technical effect of setting the material optimization fitness, performing multiple iterations in the optimization space, comparing the fitness, analyzing the acceptance probability method, determining the optimal parameters among multiple material optimization parameters, improving the accuracy of lens material optimization, realizing the autonomous optimization of collision detection, and further ensuring the application performance of resin lenses.

[0060] Furthermore, the steps of this application further include:

[0061] Step S810: Obtain the service life information of the resin lens;

[0062] Step S820: Conduct a vision impact evaluation on the service life information of the resin lens and the collision detection result of the resin lens to obtain the lens vision impact coefficient;

[0063] Step S830: Modify the material optimization guidance scheme based on the lens vision impact coefficient.

[0064] Specifically, to ensure the accuracy of the material optimization scheme, obtain the service life information of the test resin lens. The normal service life of a resin lens is generally two years, and as the service life increases, it will have an adverse impact on vision. Then conduct a vision impact evaluation on the service life information of the resin lens and the collision detection result of the resin lens. Exemplarily, scratches on the surface of the resin lens, geometric dimension structure, etc. will all have different degrees of impact on vision, and obtain its lens vision impact coefficient. The larger the coefficient, the greater the adverse impact on vision. Modify the material optimization guidance scheme based on the lens vision impact coefficient, such as increasing the use strength or hardness of the lens material. By combining the vision impact factors of the lens to modify the lens optimization scheme, the accuracy of resin lens material optimization is higher, and further the application performance of resin lenses is ensured.

[0065] In summary, a material optimization method and device based on resin lens collision detection provided by this application have the following technical effects:

[0066] By adopting the method of extracting elements from the information of the lens collision application scenario, and then performing data detection through the extracted collision scenario application parameter information to obtain the resin lens collision detection result, obtaining the lens material optimization requirement information according to the resin lens collision detection result and the resin lens material parameter information, then constructing a material optimization fitness function based on the lens material optimization requirement information, obtaining the lens material optimization space according to the resin lens material parameter information, performing optimization within the lens material optimization space based on the material optimization fitness function, outputting a material optimization guiding scheme, and performing lens material optimization management according to the material optimization guiding scheme. Furthermore, it achieves the technical effect of determining the material optimization parameter scheme through the resin lens for scene collision testing, realizing the autonomous optimization of collision detection, improving the accuracy of lens material optimization, and further ensuring the application performance of the resin lens.

[0067] Embodiment 2

[0068] Based on the same inventive concept as the material optimization method for resin lens collision detection in the foregoing embodiment, the present invention also provides a material optimization device for resin lens collision detection, as Figure 4 shown, the device includes:

[0069] A scene acquisition module 11, configured to acquire lens collision application scenario information;

[0070] An element extraction module 12, configured to extract elements from the lens collision application scenario information to obtain collision scenario application parameter information;

[0071] A data detection module 13, configured to perform data detection through the collision scenario application parameter information to obtain a resin lens collision detection result;

[0072] An optimization requirement acquisition module 14, configured to acquire resin lens material parameter information, and obtain lens material optimization requirement information according to the resin lens collision detection result and the resin lens material parameter information;

[0073] A fitness construction module 15, configured to construct a material optimization fitness function based on the lens material optimization requirement information;

[0074] An optimization space acquisition module 16, configured to obtain a lens material optimization space according to the resin lens material parameter information;

[0075] An optimization management module 17, configured to perform optimization within the lens material optimization space based on the material optimization fitness function, output a material optimization guiding scheme, and perform lens material optimization management according to the material optimization guiding scheme.

[0076] Furthermore, the data detection module further includes:

[0077] An orthogonal arrangement unit for orthogonally arranging the parameter information applied to the collision scenario to obtain a collision test scenario parameter table;

[0078] A collision test unit for performing a lens collision test based on the parameter information applied to the collision scenario to obtain lens collision detection data information and lens collision picture acquisition information;

[0079] A feature point extraction unit for extracting feature points from the lens collision picture acquisition information to obtain lens collision detection defect information;

[0080] A detection result obtaining unit for obtaining the resin lens collision detection result based on the lens collision detection data information and the lens collision detection defect information.

[0081] Furthermore, the feature point extraction unit further includes:

[0082] A model building unit for building a lens collision defect detection model, where the lens collision defect detection model includes an input layer, an image recognition layer, a defect classification layer, and an output layer;

[0083] An image recognition unit for taking the lens collision picture acquisition information as the input layer and inputting it into the image recognition layer to obtain lens collision feature information;

[0084] A defect classification unit for inputting the lens collision feature information into the defect classification layer to obtain the lens collision detection defect information;

[0085] A model output unit for outputting the lens collision detection defect information as an output result through the output layer.

[0086] Furthermore, the image recognition unit further includes:

[0087] A convolution feature obtaining unit for obtaining a predetermined image convolution feature according to the lens smoothness standard;

[0088] A feature extraction unit for inputting the lens collision picture acquisition information into the image recognition layer for feature extraction;

[0089] A feature recognition unit for obtaining the output information of the image recognition layer, where the output information is the lens collision feature information that does not conform to the predetermined image convolution feature.

[0090] Furthermore, the optimization management module further includes:

[0091] A material optimal parameter selection unit for randomly selecting material optimization parameters from the lens material optimization space as the material optimal parameters;

[0092] An optimized fitness calculation unit, which is used to optimize the fitness function according to the material and calculate the optimized fitness of the optimal parameters of the material;

[0093] A comparison parameter selection unit, which is used to randomly select material optimization parameters from the optimization space of the lens material as comparison parameters;

[0094] A comparison fitness calculation unit, which is used to calculate the comparison fitness of the comparison parameters according to the material optimized fitness function;

[0095] A parameter substitution unit, which is used to substitute the comparison parameters as the optimal parameters of the material if the comparison fitness is greater than the optimized fitness;

[0096] A parameter output unit, which is used to output the optimal parameters of the material and obtain the material optimization guidance scheme if the optimal parameters of the material do not change in the iterative optimization of the threshold number of times or the iterative optimization reaches the preset number of times.

[0097] Furthermore, the device further includes:

[0098] A probability calculation unit, which is used to substitute the comparison parameters as the optimal parameters of the material according to a probability if the comparison fitness is less than the optimized fitness, and the probability is calculated by the following formula:

[0099]

[0100] where r2 is the comparison fitness, r1 is the optimized fitness, and k is the optimization speed factor.

[0101] Furthermore, the device further includes:

[0102] A lens service life acquisition unit, which is used to acquire the resin lens service life information;

[0103] A vision impact evaluation unit, which is used to evaluate the vision impact of the resin lens service life information and the resin lens collision detection result to obtain a lens vision impact coefficient;

[0104] An optimization scheme correction unit, which is used to correct the material optimization guidance scheme based on the lens vision impact coefficient.

[0105] The present application provides a material optimization method based on resin lens collision detection. The method includes: obtaining lens collision application scenario information; extracting elements from the lens collision application scenario information to obtain collision scenario application parameter information; performing data detection through the collision scenario application parameter information to obtain a resin lens collision detection result; obtaining resin lens material parameter information, and based on the resin lens collision detection result and the resin lens material parameter information, obtaining lens material optimization requirement information; constructing a material optimization fitness function based on the lens material optimization requirement information; obtaining a lens material optimization search space according to the resin lens material parameter information; performing optimization within the lens material optimization search space based on the material optimization fitness function, outputting a material optimization guidance plan, and performing lens material optimization management according to the material optimization guidance plan. It solves the technical problem in the prior art that the optimization accuracy of material components is low, resulting in the influence on the application performance of resin lenses. It achieves the technical effect of determining a material optimization parameter scheme through scene collision testing of resin lenses, realizing the autonomous optimization of collision detection, improving the accuracy of lens material optimization, and further ensuring the application performance of resin lenses.

[0106] This specification and the drawings are merely illustrative of the present application. If the modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

1. A material optimization method based on the collision detection of resin lenses, characterized in that, The method includes: Obtaining lens collision application scenario information; Performing element extraction on the lens collision application scenario information to obtain collision scenario application parameter information; Performing data detection through the collision scenario application parameter information to obtain a resin lens collision detection result; Obtaining resin lens material parameter information, and based on the resin lens collision detection result and the resin lens material parameter information, obtaining lens material optimization requirement information; Constructing a material optimization fitness function based on the lens material optimization requirement information; Obtaining a lens material optimization search space according to the resin lens material parameter information; Performing optimization within the lens material optimization search space based on the material optimization fitness function, outputting a material optimization guidance plan, and performing lens material optimization management according to the material optimization guidance plan.

2. The method according to claim 1, wherein The performing data detection through the collision scenario application parameter information to obtain a resin lens collision detection result includes: Performing orthogonal arrangement on the collision scenario application parameter information to obtain a collision test scenario parameter table; Performing a lens collision test based on the collision scenario application parameter information to obtain lens collision detection data information and lens collision picture acquisition information; Performing feature point extraction on the lens collision picture acquisition information to obtain lens collision detection defect information; Based on the lens collision detection data information and the lens collision detection defect information, obtaining the resin lens collision detection result.

3. The method according to claim 2, wherein The performing feature point extraction on the lens collision picture acquisition information to obtain lens collision detection defect information includes: Building a lens collision defect detection model, where the lens collision defect detection model includes an input layer, an image recognition layer, a defect classification layer, and an output layer; Taking the lens collision picture acquisition information as the input layer and inputting it into the image recognition layer to obtain lens collision feature information; Inputting the lens collision feature information into the defect classification layer to obtain the lens collision detection defect information; Taking the lens collision detection defect information as the output result and outputting it through the output layer.

4. The method according to claim 3, wherein The taking the lens collision picture acquisition information as the input layer and inputting it into the image recognition layer to obtain lens collision feature information includes: Obtaining predetermined image convolution features according to the lens smoothness standard; Inputting the lens collision picture acquisition information into the image recognition layer for feature extraction; Obtaining the output information of the image recognition layer, where the output information is the lens collision feature information that does not conform to the predetermined image convolution features.

5. The method according to claim 1, wherein The performing optimization within the lens material optimization search space based on the material optimization fitness function and outputting a material optimization guidance plan includes: Randomly selecting material optimization parameters from the lens material optimization search space as the optimal material parameters; Calculating the optimization fitness of the optimal material parameters according to the material optimization fitness function; Randomly selecting material optimization parameters from the lens material optimization search space as comparison parameters; Calculating the comparison fitness of the comparison parameters according to the material optimization fitness function; If the comparison fitness is greater than the optimization fitness, the comparison parameter is replaced as the optimal material parameter; If the optimal material parameter does not change during the iterative optimization for the threshold number of times or the iterative optimization reaches the preset number of times, the optimal material parameter is output to obtain the material optimization guidance scheme.

6. The method according to claim 5, wherein The method includes: If the comparison fitness is less than the optimization fitness, the comparison parameter is replaced as the optimal material parameter with a probability, and the probability is calculated by the following formula: where r2 is the comparison fitness, r1 is the optimization fitness, and k is the optimization speed factor.

7. The method according to claim 1, wherein The method includes: Obtain the service life information of the resin lens; Conduct a vision impact evaluation on the service life information of the resin lens and the resin lens collision detection result to obtain the lens vision impact coefficient; Based on the lens vision impact coefficient, correct the material optimization guidance scheme.

8. A material optimization device based on resin lens collision detection, characterized in that, The device includes: A scene acquisition module for acquiring lens collision application scene information; A feature extraction module for extracting features from the lens collision application scene information to obtain collision scene application parameter information; A data detection module for performing data detection through the collision scene application parameter information to obtain a resin lens collision detection result; An optimization requirement acquisition module for acquiring resin lens material parameter information, and obtaining lens material optimization requirement information according to the resin lens collision detection result and the resin lens material parameter information; A fitness construction module for constructing a material optimization fitness function based on the lens material optimization requirement information; An optimization space acquisition module for obtaining a lens material optimization space according to the resin lens material parameter information; An optimization management module for performing optimization within the lens material optimization space based on the material optimization fitness function, outputting a material optimization guidance scheme, and performing lens material optimization management according to the material optimization guidance scheme.

Images