A prism lens combination optical kit and production method thereof

Abstract

The present invention discloses a prism lens combination optical kit and a production method thereof. The optical kit comprises a mounting portion, a kit structure, a prism lens combination optical element, and a production and installation detection component. The prism lens combination optical element is glued in a card slot of the kit structure, so that the optical kit can realize the functions of a prism and a lens. During the assembly and production process, by detecting whether the prism lens combination optical element is tightly glued to the card slot, light path experimental data is collected, and by determining whether the prism lens combination optical element is tightly glued to the card slot, it is determined whether the prism lens effect of the optical kit meets the standard; if the standard is not met, a prompt message is issued, and the positional relationship between the card slot and the prism lens combination optical element is adjusted in time, the design and production process are optimized, and the performance and stability of the optical element are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of optical electronic components, and in particular to a prism lens combination optical kit and a production method thereof. Background Art

[0002] Optical components play a vital role in modern technology and are widely used in cameras, microscopes, telescopes, laser equipment, and various other optical instruments. Prisms and lenses are two basic optical components, each with unique functions and applications.

[0003] Applications and Functions of Prisms: Prisms can change the direction of light propagation, deflecting incident light through the principle of refraction. They are widely used in spectral analysis, imaging systems, and optical instruments, achieving functions such as dispersion, reflection, and light deflection. In cameras and projectors, prisms can be used to correct image direction, ensuring image accuracy and clarity.

[0004] Lens properties and applications: Lenses are primarily used to focus or diverge light, changing its focal length. Convex lenses focus light and are often used for magnification and imaging, while concave lenses diffuse light and are widely used in eyewear and optical instruments. The design and manufacturing precision of lenses is crucial to their optical performance, especially in high-end optical equipment, where even the slightest error can result in blurred or distorted images.

[0005] Defects of existing optical components: Currently, optical components on the market mainly rely on gluing two molded lenses and prisms together. During the assembly process, there is often a problem of loose docking between optical components, resulting in unstable optical paths and affecting optical performance.

[0006] Despite certain inspection methods during the production process, there is still a lack of real-time monitoring and analysis of the bonding effect of optical components, which can easily lead to fluctuations in product quality. In addition, traditional optical components are usually designed and manufactured independently, failing to fully utilize the combined advantages of prisms and lenses, resulting in increased complexity of the optical system and greater difficulty in assembly.

[0007] Demand for automation and intelligence: With the development of intelligent manufacturing and automation technology, higher requirements are placed on the production process of optical components. There is an urgent need for a new type of optical kit that can quickly and accurately achieve an effective combination of prisms and lenses so that the optical components of the kit have the optical properties of prisms and lenses.

[0008] By introducing data acquisition and analysis technology, the performance of optical components can be monitored in real time to ensure that each optical component meets the design standards, thereby improving production efficiency and product reliability.

[0009] In summary, existing prism and lens optical assemblies have significant deficiencies in functional integration, assembly precision, and quality control. To address these issues, the present invention proposes a novel prism-lens combination optical kit and its production method, aiming to improve the performance and stability of optical components. Summary of the Invention

[0010] The object of the present invention is to provide a prism lens combination optical kit and a production method thereof, so as to solve the above-mentioned problems existing in the prior art.

[0011] In a first aspect, the present invention provides a prism lens combination optical kit, including a production and installation detection component, wherein the production and installation detection component includes a processor, a camera, and an optical path detection component;

[0012] The camera is used to capture a cross-sectional image of the contact surface between the prism lens combination optical element and the kit structure, and send the cross-sectional image to the processor; the light path detection component is used to obtain light path data when the prism lens combination optical element is installed in the card slot, and send the light path data to the processor;

[0013] The processor is configured to determine whether the prism lens combination optical element is tightly glued to the card slot based on the cross-sectional image and the optical path data, and determine whether the prism lens effect of the optical kit meets the standard; if not, issue a prompt message;

[0014] Wherein, judging whether the prism lens combination optical element is tightly glued to the card slot according to the cross-sectional image and the optical path data includes:

[0015] Preprocessing: Denoise and enhance the cross-sectional image to obtain a preprocessed image; clean the optical path data to obtain cleaned optical path data;

[0016] Extracting contact surface contours based on preprocessed images;

[0017] The contact area and contact uniformity index are obtained based on the contact surface profile and the cleaning optical path data; the contact uniformity index is used to characterize the degree of contact uniformity between the prism lens combination optical element and the card slot;

[0018] Based on the contact area and the contact uniformity index, a tight bonding index is obtained; the tight bonding index is used to indicate whether the prism lens combination optical element and the card slot are tightly bonded;

[0019] If the tight bonding index is less than the set value, it is determined that the prism lens combination optical element is not tightly bonded to the card slot;

[0020] If the tight bonding index is greater than or equal to the set value, it is determined that the prism lens combination optical element is tightly bonded to the card slot.

[0021] Optionally, obtaining the contact area and the contact uniformity index based on the contact surface profile and the cleaning light path data includes:

[0022] Constructing a light path image based on the light path data; extracting the contact area between the prism lens combination optical element and the card slot based on the light path image;

[0023] Obtaining the contact area based on the contact area;

[0024] Extracting light intensity information of the contact area from the cleaning optical path data to obtain a first-dimensional contact vector; the first-dimensional contact vector represents the contact condition between the mirror assembly optical element and the card slot in the first dimension, and the first-dimensional contact vector is represented by a light intensity distribution vector;

[0025] A second-dimensional contact vector is obtained based on the contact surface profile; the second-dimensional contact vector represents the contact condition between the mirror assembly optical element and the card slot in the second dimension;

[0026] The contact uniformity index is obtained by combined analysis based on the first-dimensional contact vector and the second-dimensional contact vector.

[0027] Optionally, constructing a light path image based on the light path data includes:

[0028] The light path image is constructed by taking the light intensity of the light path data as the pixel value of the light path image and taking the light path illumination point as the coordinate point of the light path image.

[0029] Optionally, a combined analysis is performed based on the first-dimensional contact vector and the second-dimensional contact vector to obtain a contact uniformity index, including:

[0030] The first-dimensional contact vector and the second-dimensional contact vector are combined and analyzed by a joint multi-layer perceptron model, and the joint multi-layer perceptron model outputs a contact uniformity index; wherein, the layer perceptron model includes a first branch, a second branch and an interaction layer, the first branch includes multiple layers of first-branch hidden layers connected in sequence, and the second branch includes multiple layers of second-branch hidden layers connected in sequence; the interaction layer is arranged between the corresponding first-branch hidden layer and the second-branch hidden layer, connecting the first-branch hidden layer and the second-branch hidden layer; the hidden layer is used to realize information interaction between the first-branch hidden layer and the second-branch hidden layer, so as to mine the implicit information related to each other between the first-dimensional contact vector and the second-dimensional contact vector.

[0031] Optionally, the first dimension contact vector and the second dimension contact vector are combined and analyzed by a joint multilayer perceptron model, and the joint multilayer perceptron model outputs a contact uniformity index, including:

[0032] Step S1: The first branch hidden layer of the i-th layer extracts features from the first input of the i-th layer to obtain the first feature H of the i-th layer AThe second branch hidden layer of the i-th layer extracts features from the second input of the i-th layer and obtains the second feature H of the i-th layer. B ; i is a positive integer less than or equal to N, N is the number of hidden layers; N is a positive integer greater than 3; when i = 1, the first input of the i-th layer is the first-dimensional contact vector; the second input of the i-th layer is the second-dimensional contact vector;

[0033] Step S2: Feature mapping is performed in the following manner:

[0034] H′ A =f(W A H A +b A )

[0035] H′ B =f(W B H B +b B )

[0036] Among them, W A and W B are the first weight matrix learned by the first branch and the second weight matrix learned by the second branch respectively; b A and b B are the first bias term of the first branch and the second bias term of the second branch respectively; H′ A is the first mapping feature of the i-th layer, H′ B is the second mapping feature of the i-th layer, f() is the mapping function, f(W A H A +b A ) represents the mapping function for the input (W A H A +b A )’s mapping output, f(W B H B +b B ) represents the mapping function for the input (W B H B +b B )’s mapping output;

[0037] Step S3: After the interaction layer, the first mapping feature of the i-th layer and the second mapping feature of the i-th layer are combined using a fusion mechanism to obtain the i-th layer fusion feature H. The specific fusion method is shown in the following formula:

[0038] H=AH′ B +(1-A)H′ A

[0039] Where A is the attention weight matrix, which is obtained by training the joint multi-layer perceptron model; Represents H′ A The transposed matrix of , softmax() is the softmax function, Indicates that the softmax function is for input Output;

[0040] Step S4: Obtain the first feature H of the i-th layer A The first loss function of the i-th layer between the i-th layer fusion feature H; obtain the i-th layer second feature H B The second loss function of the i-th layer between the i-th layer fusion feature H;

[0041] Step S5: adjusting the fusion feature H of the i-th layer based on the first loss function of the i-th layer to obtain the first branch adjustment feature of the i-th layer; adjusting the fusion feature H of the i-th layer based on the second loss function of the i-th layer to obtain the second branch adjustment feature of the i-th layer;

[0042] Step S6: For the i+1th layer, the first branch adjusted features of the i-th layer and the second branch adjusted features of the i-th layer are used as the inputs of the first branch hidden layer of the i+1th layer and the second branch hidden layer of the i+1th layer, respectively, and the first branch adjusted features of the i+1th layer and the second branch adjusted features of the i+1th layer are obtained according to the method of steps S1 to S5;

[0043] The method shown in steps S1 to S6 is repeated until i=N, and the last first-branch hidden layer and the last second-branch hidden layer respectively output the last first-branch adjusted features and the last second-branch adjusted features;

[0044] Step S7: Based on the last layer of first branch adjustment features and the last layer of second branch adjustment features, the contact uniformity index is obtained by the following formula:

[0045] Where U represents the contact uniformity index, σ(I A ) represents the standard deviation of the first branch adjustment feature of the next layer, I A Indicates the adjustment feature of the first branch of the next layer, μ(I A ) represents the mean of the adjusted features of the first branch of the next layer, ε represents an ultra-micro number, ε=0.001; σ(I B ) represents the standard deviation of the first branch adjustment feature of the next layer, I B Indicates the adjustment feature of the first branch of the next layer, μ(I B ) represents the mean of the adjusted features of the first branch of the next layer; λ and ξ represent the first weighting factor and the second weighting factor, respectively, and the first weighting factor and the second weighting factor are the eigenvalues ​​of the attention weight matrix, respectively.

[0046] Optionally, if the eigenvalues ​​of the attention weight matrix are all 0 or there are no real eigenvalues, then λ = 0 and ξ = 1.

[0047] Optionally, obtaining a tight bonding index based on the contact area and the contact uniformity index includes:

[0048] Obtain the projection area of ​​the first-dimensional contact vector mapped into the optical path image;

[0049] The tight bonding index is obtained based on the contact uniformity index and the ratio K of the projected area to the contact area.

[0050] Optionally, based on the contact uniformity index and the ratio K of the projected area to the contact area, a tight bonding index is obtained, including:

[0051] The contact uniformity index and the ratio K of the projected area to the contact area are weighted summed to obtain the tight bonding index, which is specifically obtained by the following formula:

[0052] D=ξλU+(1-λξ)K

[0053] Where D is the tight bonding index.

[0054] Optionally, the prism lens combination optical kit further includes the following components:

[0055] Mounting part: a structure used to fix the optical kit;

[0056] Kit structure: a frame for providing support and connection, with a card slot provided on one side of the kit structure;

[0057] Prism lens combination optical element: the prism lens combination optical element is an optical element of an integrally formed prism lens combination;

[0058] The prism-lens combination optical element is glued into the card slot of the kit structure, so that the prism-lens combination optical kit can realize the functions of prism and lens; the kit structure is connected to the mounting part on the other side away from the card slot; the mounting part is used to install the kit structure and the prism-lens combination optical element on the equipment; the installation detection component is arranged at a position of the kit structure away from the mounting part.

[0059] An embodiment of the present invention further provides a method for producing a prism-lens combination optical kit, which is applied to the process of producing the prism-lens combination optical kit. The prism-lens combination optical kit includes a production and installation detection component, a mounting portion, a kit structure, and a prism-lens combination optical element.

[0060] Mounting part: a structure used to fix the optical kit;

[0061] Kit structure: a frame for providing support and connection, with a card slot provided on one side of the kit structure;

[0062] Prism lens combination optical element: the prism lens combination optical element is an optical element of an integrally formed prism lens combination;

[0063] The prism-lens combination optical element is glued into the card slot of the kit structure, so that the prism-lens combination optical kit can realize the functions of a prism and a lens; the kit structure is connected to the mounting portion on the other side away from the card slot; the mounting portion is used to mount the kit structure and the prism-lens combination optical element on the device; the installation detection component is arranged at a position of the kit structure away from the mounting portion;

[0064] Production and installation of detection components, including processors, cameras, and optical path detection components;

[0065] A method for producing a prism lens combination optical kit, comprising:

[0066] Capturing a cross-sectional image of a contact surface between the prism-lens combination optical element and the kit structure through a camera, and sending the cross-sectional image to a processor; obtaining optical path data of the prism-lens combination optical element when it is installed in the card slot through an optical path detection component, and sending the optical path data to the processor;

[0067] The processor determines whether the prism lens combination optical element is tightly glued to the card slot based on the cross-sectional image and the optical path data, and determines whether the prism lens effect of the optical kit meets the standard; if not, a prompt message is issued;

[0068] Wherein, judging whether the prism lens combination optical element is tightly glued to the card slot according to the cross-sectional image and the optical path data includes:

[0069] Preprocessing: Denoise and enhance the cross-sectional image to obtain a preprocessed image; clean the optical path data to obtain cleaned optical path data;

[0070] Extracting contact surface contours based on preprocessed images;

[0071] The contact area and contact uniformity index are obtained based on the contact surface profile and the cleaning optical path data; the contact uniformity index is used to characterize the degree of contact uniformity between the prism lens combination optical element and the card slot;

[0072] Based on the contact area and the contact uniformity index, a tight bonding index is obtained; the tight bonding index is used to indicate whether the prism lens combination optical element and the card slot are tightly bonded;

[0073] If the tight bonding index is less than the set value, it is determined that the prism lens combination optical element is not tightly bonded to the card slot;

[0074] If the tight bonding index is greater than or equal to the set value, it is determined that the prism lens combination optical element is tightly bonded to the card slot.

[0075] Compared with the prior art, the embodiments of the present invention achieve the following beneficial effects:

[0076] An embodiment of the present invention provides a prism lens combination optical kit and a production method thereof. A camera is used to capture a cross-sectional image of the contact surface between the prism lens combination optical element and the kit structure, and the cross-sectional image is sent to a processor. An optical path detection component is used to obtain optical path data of the prism lens combination optical element when it is installed in a card slot, and the optical path data is sent to the processor. The processor determines whether the prism lens combination optical element is tightly glued to the card slot based on the cross-sectional image and the optical path data, and determines whether the prism lens effect of the optical kit meets the requirements. If not, a prompt message is issued. Determining whether the prism lens combination optical element is tightly glued to the card slot based on the cross-sectional image and the optical path data includes:

[0077] Preprocessing: De-noise and enhance the cross-sectional image to obtain a preprocessed image; clean the optical path data to obtain cleaned optical path data; extract the contact surface profile based on the preprocessed image; obtain the contact area and contact uniformity index based on the contact surface profile and the cleaned optical path data; the contact uniformity index is used to characterize the degree of contact uniformity between the prism lens combination optical element and the card slot; based on the contact area and the contact uniformity index, obtain the tight bonding index; the tight bonding index is used to characterize whether the prism lens combination optical element and the card slot are tightly bonded; if the tight bonding index is less than the set value, it is determined that the prism lens combination optical element is not tightly bonded to the card slot; if the tight bonding index is greater than or equal to the set value, it is determined that the prism lens combination optical element is tightly bonded to the card slot.

[0078] By adopting the above solution, the accuracy and reliability of determining whether the optical components of the mirror assembly are tightly bonded to the card slot can be improved, thereby improving the accuracy of determining whether the prism lens effect of the optical kit meets the requirements. On this basis, the design and production process can be optimized, and the performance and stability of the optical components can be improved. In detail, the embodiments of the present invention achieve the following beneficial effects:

[0079] High-precision optical performance: This optical kit combines optical components by precisely cementing prism lenses to ensure precise alignment of the light path, thereby improving optical performance, enhancing imaging quality, and reducing optical aberration and distortion.

[0080] Flexible functional integration: The optical kit combines the functions of prisms and lenses to achieve a variety of optical effects to meet the needs of different application scenarios, such as applications in microscopes, projectors, and optical sensors.

[0081] Simplified production processes: By inspecting the tight fit of optical components in the slots during production, problems can be promptly identified and corrected, streamlining the production process and improving efficiency. Data-driven quality control: Collecting and analyzing optical path experimental data provides a basis for quality control of optical kits, ensuring that every product meets design standards and improving product consistency and reliability.

[0082] Adjustability and modularity: By adjusting the positional relationship between the card slot and the optical element, fine-tuning can be easily performed to adapt to different application requirements or compensate for manufacturing errors, increasing the adaptability of the optical kit.

[0083] Reduce production costs: This method can reduce rework and scrap rates due to unqualified optical components, thereby reducing overall production costs and improving economic benefits.

[0084] Enhance product competitiveness: High-quality and high-stability optical kits will enhance the market competitiveness of products, meet the needs of the high-end market, and attract more customers.

[0085] Easy maintenance and replacement: The modular design makes the maintenance and replacement of optical components more convenient, extending the service life of the optical kit and reducing maintenance costs.

[0086] Widely adaptable application scope: This optical kit can be widely used in scientific research, medical treatment, industrial detection, optical communication and other fields, and has good market prospects.

[0087] Promote the development of optical technology: Through innovative production methods and efficient quality control, promote the research and development and application of optical components, and contribute to the advancement of optical technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 The present invention provides a flowchart of a method for producing a prism lens combination optical kit. DETAILED DESCRIPTION

[0089] With the development of intelligent manufacturing and automation technology, higher requirements are being placed on the production process of optical components. There is an urgent need for a new type of optical kit that can quickly and accurately achieve the effective combination of prisms and lenses. This application introduces data acquisition and analysis technology to monitor the performance of optical components in real time, ensuring that each optical component meets the design standards, improving production efficiency and product reliability.

[0090] The present invention will be described in detail below with reference to the accompanying drawings.

[0091] Example 1

[0092] An embodiment of the present invention provides a prism lens combination optical kit, including a production and installation detection component, which includes a processor, a camera, and an optical path detection component.

[0093] Among them, the camera is used to capture a cross-sectional image of the contact surface between the prism lens combination optical element and the kit structure, and send the cross-sectional image to the processor. The optical path detection component is used to obtain optical path data when the prism lens combination optical element is installed on the card slot, and send the optical path data to the processor. The processor is used to determine whether the prism lens combination optical element is tightly glued to the card slot based on the cross-sectional image and the optical path data, and determine whether the prism lens effect of the optical kit meets the standards; if not, a prompt message is issued. In an embodiment of the present invention, the prompt message includes reinstalling the prism lens combination optical element or reproducing, adjusting the kit structure, and issuing information that the quality of the prism lens combination optical original does not meet the standards.

[0094] In an embodiment of the present invention, judging whether the prism-lens combination optical element is tightly glued to the card slot according to the cross-sectional image and the optical path data includes:

[0095] Preprocessing: Denoise and enhance the cross-sectional image to obtain a preprocessed image; clean the optical path data to obtain cleaned optical path data.

[0096] In an embodiment of the present invention, a high-resolution camera is used to capture a cross-sectional image of a prism-lens combination optical element, ensuring that the image is captured under good lighting conditions to obtain a clear image. A laser or other light source is used to measure the optical path data through the optical element, including the incident direction, exit direction, light intensity distribution, light intensity, and phase information.

[0097] During preprocessing, the acquired images undergo denoising and contrast enhancement to improve the accuracy of subsequent analysis. Denoising can be performed using methods such as mean filtering, median filtering, and Gaussian filtering. Enhancement can be performed using histogram equalization, brightness and contrast adjustment, Laplacian filtering, and high-pass filtering to enhance image edges and details.

[0098] Cleaning the optical path data: Recording the light intensity distribution in the contact area and measuring it using an optical sensor or camera. Data integration: Ensuring spatial alignment between the optical path data and the image data requires coordinate transformation or interpolation.

[0099] After preprocessing, the contact surface contour is extracted based on the preprocessed image. In this embodiment of the present invention, the contact surface contour can be extracted using a Canny operator. Alternatively, an edge detection algorithm can be used to extract the contour of the contact area in the processed image. Morphological operations (such as dilation and erosion) can be used to further clean the contour to obtain the final contact surface contour.

[0100] In order to improve the accuracy, reliability and convenience of judging whether the prism lens combination optical element is tightly glued to the card slot, the following method is used:

[0101] The contact area and contact uniformity index are obtained based on the contact surface profile and cleaning optical path data. The contact uniformity index is used to characterize the degree of contact uniformity between the prism lens assembly optical element and the card slot.

[0102] As a further illustration, the contact area and contact uniformity index are obtained based on the contact surface profile and cleaning optical path data, including:

[0103] A light path image is constructed based on the light path data, and the contact area between the prism lens assembly optical element and the card slot is extracted based on the light path image. Specifically, the light path image is constructed based on the light path data by using the light intensity of the light path data as the pixel value of the light path image and the light path illumination point as the coordinate point of the light path image.

[0104] The contact area is obtained based on the contact region. Specifically, the contact region can be detected using the Canny operator or the Huffman operator. The contact region is then binarized, with white representing the contact region and black representing the non-contact region. The contact area is obtained by multiplying the number of white pixels in the binary image by the area occupied by each pixel.

[0105] Light intensity information of the contact area is extracted from the cleaning optical path data to obtain a first-dimensional contact vector. The first-dimensional contact vector represents the contact between the mirror assembly optical element and the card slot in the first dimension, and the first-dimensional contact vector is represented by a light intensity distribution vector. Specifically, the light intensity of the cleaning optical path data is mapped to the contact area, and then feature extraction is performed on the contact area to obtain the first-dimensional contact vector. Feature extraction can be performed using a convolutional neural network or principal component analysis to extract features and obtain the first-dimensional contact vector. In this embodiment of the present invention, the first dimension is information in the coordinate axis X direction, or dimensional information of a plane passing through the X axis, and the second dimension is information in the coordinate axis Y direction, or dimensional information on a plane perpendicular to the plane of the first dimension and passing through the Y axis. That is, the first-dimensional contact vector is feature information observed in the X direction, that is, the front of the contact surface is observed, and the second-dimensional contact vector is feature information observed in the Y direction, that is, the cross-section of the contact surface is observed.

[0106] A second-dimensional contact vector is obtained based on the contact surface profile. The second-dimensional contact vector represents the contact between the optical element of the mirror assembly and the slot in the second dimension. Specifically, the second-dimensional contact vector is extracted from the contact surface profile to obtain the second-dimensional contact vector. A convolutional neural network can be used for feature extraction to obtain the second-dimensional contact vector.

[0107] Based on the combined analysis of the first dimension contact vector and the second dimension contact vector, the contact uniformity index is obtained. Specifically:

[0108] The first-dimensional contact vector and the second-dimensional contact vector are combined and analyzed using a multi-layer perceptron model, which then outputs a contact uniformity index. The multi-layer perceptron model comprises a first branch, a second branch, and an interaction layer. The first branch comprises multiple layers of sequentially connected first-branch hidden layers, and the second branch comprises multiple layers of sequentially connected second-branch hidden layers. The interaction layer is positioned between the corresponding first-branch hidden layers and the second-branch hidden layers, connecting them. The hidden layer is used to implement information exchange between the first-branch hidden layers and the second-branch hidden layers, thereby mining the implicit information associated with the first-dimensional contact vector and the second-dimensional contact vector.

[0109] In an embodiment of the present invention, a combined analysis is performed on the first-dimensional contact vector and the second-dimensional contact vector by using a joint multi-layer perceptron model, and the joint multi-layer perceptron model outputs a contact uniformity index, including:

[0110] Step S1: The first branch hidden layer of the i-th layer extracts features from the first input of the i-th layer to obtain the first feature H of the i-th layer A The second branch hidden layer of the i-th layer extracts features from the second input of the i-th layer and obtains the second feature H of the i-th layer. B i is a positive integer less than or equal to N, where N is the number of hidden layers. N is a positive integer greater than 3. When i = 1, the first input of the i-th layer is the first-dimensional contact vector; the second input of the i-th layer is the second-dimensional contact vector. That is, the first-dimensional contact vector is used as the input of the first-branch hidden layer of the first layer, and the second-dimensional contact vector is used as the input of the second-branch hidden layer of the second layer. The first-branch hidden layer and the first-branch hidden layer are used for feature extraction respectively.

[0111] Step S2: Feature mapping is performed in the following manner:

[0112] H′ A =f(W A H A +b A )

[0113] H′ B =f(W BH B +b B )

[0114] Among them, W A and W B are the first weight matrix learned by the first branch and the second weight matrix learned by the second branch respectively; b A and b B are the first bias term of the first branch and the second bias term of the second branch respectively; H′ A is the first mapping feature of the i-th layer, H′ B is the second mapping feature of the i-th layer, f() is the mapping function, which can be a ReLU function or a Swish function. f(W A H A +b A ) represents the mapping function for the input (W A H A +b A )’s mapping output, f(W B H B +b B ) represents the mapping function for the input (W B H B +b B )’s mapping output.

[0115] Step S3: After the interaction layer, the first mapping feature of the i-th layer and the second mapping feature of the i-th layer are combined using a fusion mechanism to obtain the i-th layer fusion feature H. The specific fusion method is shown in the following formula:

[0116] H=AH′ B +(1-A)H′ A

[0117] Where A is the attention weight matrix, which is obtained by training the joint multi-layer perceptron model; A = Represents H′ A The transposed matrix of , softmax() is the softmax function, Indicates that the softmax function is for input Output.

[0118] Step S4: Obtain the first feature H of the i-th layer A The first loss function of the i-th layer between the i-th layer fusion feature H, the i-th layer first loss function can be the i-th layer first feature H A Cross entropy between the fusion feature H of the i-th layer; obtain the second feature H of the i-th layer B The second loss function of the i-th layer between the i-th layer fusion feature H. The second loss function of the i-th layer can be the second feature H of the i-th layer BThe cross entropy between the fusion feature H of the i-th layer.

[0119] Step S5: Adjust the fusion feature H of the i-th layer based on the first loss function of the i-th layer to obtain the first branch adjustment feature of the i-th layer; adjust the fusion feature H of the i-th layer based on the second loss function of the i-th layer to obtain the second branch adjustment feature of the i-th layer.

[0120] Step S6: For the i+1th layer, the first branch adjustment feature of the i-th layer and the second branch adjustment feature of the i-th layer are used as the input of the first branch hidden layer of the i+1th layer and the second branch hidden layer of the i+1th layer, respectively, and the first branch adjustment feature of the i+1th layer and the second branch adjustment feature of the i+1th layer are obtained according to the method of steps S1 to S5.

[0121] The method shown in steps S1 to S6 is looped until i=N, and the last first branch hidden layer and the last second branch hidden layer respectively output the last first branch adjustment feature and the last second branch adjustment feature.

[0122] Step S7: Obtain a contact uniformity index based on the last layer first branch adjustment feature and the last layer second branch adjustment feature using the following formula.

[0123] Where U represents the contact uniformity index, σ(I A ) represents the standard deviation of the first branch adjustment feature of the next layer, I A Indicates the adjustment feature of the first branch of the next layer, μ(I A ) represents the mean of the adjusted features of the first branch of the next layer, ε represents an ultra-micro number, ε=0.001; σ(I B ) represents the standard deviation of the first branch adjustment feature of the next layer, I B Indicates the adjustment feature of the first branch of the next layer, μ(I B ) represents the mean of the adjusted features of the first branch of the next layer; λ and ξ represent the first weighting factor and the second weighting factor, respectively, and the first weighting factor and the second weighting factor are the eigenvalues ​​of the attention weight matrix, respectively.

[0124] By adopting the above scheme, the innovative joint multi-layer perceptron (MLP) model structure, through multiple input branches, feature fusion layers, and attention mechanisms, can effectively capture the complex correlations between them when processing two sets of data with different dimensions, directions, and meanings. This design not only improves the model's expressiveness, but also enhances its flexibility and interpretability. At the same time, by introducing an interaction layer and fusion mechanism, the first and second branches can effectively share and fuse each other's implicit information, thereby enhancing the model's ability to learn complex data relationships and can spread the implicit correlation between the first-dimensional contact vector and the second-dimensional contact vector. At the same time, the contact uniformity index obtained in this way can accurately, quickly, and effectively characterize the uniformity of the bonding between the mirror assembly optical element and the card slot, thereby improving the accuracy and reliability of judging whether the mirror assembly optical element and the card slot are tightly bonded.

[0125] In an embodiment of the present invention, if the eigenvalues ​​of the attention weight matrix are all 0 or there is no real eigenvalue, then λ=0 and ξ=1.

[0126] After obtaining the contact uniformity index, the tight bonding index is obtained based on the contact area and the contact uniformity index; the tight bonding index is used to characterize whether the prism lens combination optical element and the card slot are tightly bonded.

[0127] Optionally, obtaining a tight bonding index based on the contact area and the contact uniformity index includes:

[0128] Obtain the projected area of ​​the first-dimensional contact vector mapped onto the optical path image. Obtain a tight bonding index based on the contact uniformity index and the ratio K of the projected area to the contact area. Specifically, a weighted summation of the contact uniformity index and the ratio K of the projected area to the contact area is performed to obtain the tight bonding index, which is specifically obtained using the following formula:

[0129] D=ξλU+(1-λξ)K

[0130] Where D is the tight bonding index.

[0131] In this way, the obtained tight bonding index can accurately represent the bonding degree between the prism lens combination optical element and the card slot.

[0132] Furthermore, if the tight gluing index is less than a set value, it is determined that the prism lens combination optical element is not tightly glued to the card slot; if the tight gluing index is greater than or equal to the set value, it is determined that the prism lens combination optical element is tightly glued to the card slot. In an embodiment of the present invention, the set value is between 0.05 and 20.

[0133] By adopting the above solution, the accuracy and reliability of judging whether the mirror assembly optical element and the card slot are tightly glued can be improved.

[0134] As an optional embodiment, the prism lens combination optical kit includes the following components in addition to the above-mentioned production and installation detection components:

[0135] Mounting part: The structure used to fix the optical kit.

[0136] The kit structure is a frame for providing support and connection, and a card slot is provided on one side of the kit structure.

[0137] Prism lens combination optical element: The prism lens combination optical element is an optical element of an integrally formed prism lens combination.

[0138] The prism-lens combination optical element is glued into the card slot of the kit structure, so that the prism-lens combination optical kit can realize the functions of prism and lens; the kit structure is connected to the mounting part on the other side away from the card slot; the mounting part is used to install the kit structure and the prism-lens combination optical element on the equipment; the installation detection component is arranged at a position of the kit structure away from the mounting part.

[0139] In an embodiment of the present invention, a method for producing a prism lens combination optical kit is also provided, which is used to produce the above-mentioned prism lens combination optical kit process. Figure 1 As shown, the production method of the prism lens combination optical kit includes:

[0140] S101: Capture a cross-sectional image of the contact surface between the prism lens combination optical element and the kit structure through a camera, send the cross-sectional image to a processor, obtain optical path data of the prism lens combination optical element when it is installed on the card slot through an optical path detection component, and send the optical path data to the processor.

[0141] S102: The processor determines whether the prism lens combination optical element is tightly glued to the card slot based on the cross-sectional image and the optical path data, and determines whether the prism lens effect of the optical kit meets the standard; if not, a prompt message is issued.

[0142] In the embodiment of the present invention, the specific implementation process of step S102 is described above and will not be repeated here.

[0143] By adopting the above solution, the existing prism and lens optical assemblies have obvious deficiencies in functional integration, assembly accuracy, and quality control. To address these issues, the present invention proposes a prism-lens combination optical kit and its production method, aiming to improve the performance and stability of optical components by optimizing the design and production process. Specifically, it can effectively determine whether the prism-lens combination optical component is tightly glued to the card slot and evaluate whether its optical effect meets the standard. This process combines image processing and optical path data analysis technology, which can provide strong support for the quality control of optical components.

[0144] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other device. Various general-purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing this type of system. In addition, the present invention is not directed to any specific programming language. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the above description of specific languages ​​is for the purpose of disclosing the best mode of the present invention.

[0145] In the description provided herein, numerous specific details are described. However, it is understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0146] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Accordingly, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of the invention.

[0147] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition may be divided into multiple submodules or subunits or subcomponents. All features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed herein may be combined in any combination, except that at least some of such features and / or processes or units are mutually exclusive. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0148] Furthermore, those skilled in the art will appreciate that although some embodiments herein include certain features included in other embodiments but not other features, combinations of features from different embodiments are intended to be within the scope of the present invention and to form different embodiments. For example, in the claims below, any of the claimed embodiments may be used in any combination.

[0149] The various component embodiments of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components in the apparatus according to an embodiment of the present invention. The present invention can also be implemented as a device or device program (e.g., a computer program and a computer program product) for executing a part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0150] It should be noted that the above embodiments illustrate rather than limit the invention, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between brackets should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising several different elements and by means of appropriately programmed computers. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names.

Claims

1. A prism lens combination optical kit, characterized in that: Including production and installation of detection components, said production and installation of detection components including processors, cameras and optical path detection components; The camera is used to capture a cross-sectional image of the contact surface between the prism lens combination optical element and the kit structure, and send the cross-sectional image to the processor; the light path detection component is used to obtain light path data when the prism lens combination optical element is installed in the card slot, and send the light path data to the processor; The processor is configured to determine whether the prism lens combination optical element is tightly glued to the card slot based on the cross-sectional image and the optical path data, and determine whether the prism lens effect of the optical kit meets the standard; if not, issue a prompt message; Wherein, judging whether the prism lens combination optical element is tightly glued to the card slot according to the cross-sectional image and the optical path data includes: Preprocessing: Denoise and enhance the cross-sectional image to obtain a preprocessed image; clean the optical path data to obtain cleaned optical path data; Extracting contact surface contours based on preprocessed images; The contact area and contact uniformity index are obtained based on the contact surface profile and the cleaning optical path data; the contact uniformity index is used to characterize the degree of contact uniformity between the prism lens combination optical element and the card slot; Based on the contact area and the contact uniformity index, a tight bonding index is obtained; the tight bonding index is used to indicate whether the prism lens combination optical element and the card slot are tightly bonded; If the tight bonding index is less than the set value, it is determined that the prism lens combination optical element is not tightly bonded to the card slot; If the tight bonding index is greater than or equal to the set value, it is determined that the prism lens combination optical element is tightly bonded to the card slot.

2. The prism lens combination optical kit according to claim 1, characterized in that: The method of obtaining the contact area and the contact uniformity index based on the contact surface profile and the cleaning light path data includes: Constructing a light path image based on the light path data; extracting the contact area between the prism lens combination optical element and the card slot based on the light path image; Obtaining the contact area based on the contact area; Extracting light intensity information of the contact area from the cleaning optical path data to obtain a first-dimensional contact vector; the first-dimensional contact vector represents the contact condition between the mirror assembly optical element and the card slot in the first dimension, and the first-dimensional contact vector is represented by a light intensity distribution vector; A second-dimensional contact vector is obtained based on the contact surface profile; the second-dimensional contact vector represents the contact condition between the mirror assembly optical element and the card slot in the second dimension; The contact uniformity index is obtained by combined analysis based on the first-dimensional contact vector and the second-dimensional contact vector.

3. The prism lens combination optical kit according to claim 2, wherein: The constructing of the light path image based on the light path data includes: The light path image is constructed by taking the light intensity of the light path data as the pixel value of the light path image and taking the light path illumination point as the coordinate point of the light path image.

4. The prism lens combination optical kit according to claim 3, characterized in that: Based on the combined analysis of the first-dimensional contact vector and the second-dimensional contact vector, the contact uniformity index is obtained, including: The first-dimensional contact vector and the second-dimensional contact vector are combined and analyzed by a joint multi-layer perceptron model, and the joint multi-layer perceptron model outputs a contact uniformity index; wherein, the layer perceptron model includes a first branch, a second branch and an interaction layer, the first branch includes multiple layers of first-branch hidden layers connected in sequence, and the second branch includes multiple layers of second-branch hidden layers connected in sequence; the interaction layer is arranged between the corresponding first-branch hidden layer and the second-branch hidden layer, connecting the first-branch hidden layer and the second-branch hidden layer; the hidden layer is used to realize information interaction between the first-branch hidden layer and the second-branch hidden layer, so as to mine the implicit information related to each other between the first-dimensional contact vector and the second-dimensional contact vector.

5. The prism lens combination optical kit according to claim 4, characterized in that: The first dimension contact vector and the second dimension contact vector are combined and analyzed by a joint multi-layer perceptron model, and the joint multi-layer perceptron model outputs a contact uniformity index, including: Step S1: The first branch hidden layer of the i-th layer extracts features from the first input of the i-th layer to obtain the first feature H of the i-th layer A The second branch hidden layer of the i-th layer extracts features from the second input of the i-th layer and obtains the second feature H of the i-th layer. B ; i is a positive integer less than or equal to N, N is the number of hidden layers; N is a positive integer greater than 3; when i = 1, the first input of the i-th layer is the first-dimensional contact vector; the second input of the i-th layer is the second-dimensional contact vector; Step S2: Feature mapping is performed in the following manner: H′ A =f(W A H A +b A ) H′ B =f(W B H B +b B ) Among them, W A and W B are the first weight matrix learned by the first branch and the second weight matrix learned by the second branch respectively; b A and b B are the first bias term of the first branch and the second bias term of the second branch respectively; H′ A is the first mapping feature of the i-th layer, H′ B is the second mapping feature of the i-th layer, f() is the mapping function, f(W A H A +b A ) represents the mapping function for the input (W A H A +b A )’s mapping output, f(W B H B +b B ) represents the mapping function for the input (W B H B +b B )’s mapping output; Step S3: After the interaction layer, the first mapping feature of the i-th layer and the second mapping feature of the i-th layer are combined using a fusion mechanism to obtain the i-th layer fusion feature H. The specific fusion method is shown in the following formula: H=AH′ B +(1-A)H′ A Where A is the attention weight matrix, which is obtained by training the joint multi-layer perceptron model; Represents H′ A The transposed matrix of , softmax() is the softmax function, Indicates that the softmax function is for input Output; Step S4: Obtain the first feature H of the i-th layer A The first loss function of the i-th layer between the i-th layer fusion feature H; obtain the i-th layer second feature H B The second loss function of the i-th layer between the i-th layer fusion feature H; Step S5: adjusting the fusion feature H of the i-th layer based on the first loss function of the i-th layer to obtain the first branch adjustment feature of the i-th layer; adjusting the fusion feature H of the i-th layer based on the second loss function of the i-th layer to obtain the second branch adjustment feature of the i-th layer; Step S6: For the i+1th layer, the first branch adjusted features of the i-th layer and the second branch adjusted features of the i-th layer are used as the inputs of the first branch hidden layer of the i+1th layer and the second branch hidden layer of the i+1th layer, respectively, and the first branch adjusted features of the i+1th layer and the second branch adjusted features of the i+1th layer are obtained according to the method of steps S1 to S5; The method shown in steps S1 to S6 is repeated until i=N, and the last first-branch hidden layer and the last second-branch hidden layer respectively output the last first-branch adjusted features and the last second-branch adjusted features; Step S7: Based on the last layer of first branch adjustment features and the last layer of second branch adjustment features, the contact uniformity index is obtained by the following formula: Where U represents the contact uniformity index, σ(I A ) represents the standard deviation of the first branch adjustment feature of the next layer, I A Indicates the adjustment feature of the first branch of the next layer, μ(I A ) represents the mean of the adjusted features of the first branch of the next layer, ε represents an ultra-micro number, ε=0.001; σ(I B ) represents the standard deviation of the first branch adjustment feature of the next layer, I B Indicates the adjustment feature of the first branch of the next layer, μ(I B ) represents the mean of the adjusted features of the first branch of the next layer; λ and ξ represent the first weighting factor and the second weighting factor, respectively, and the first weighting factor and the second weighting factor are the eigenvalues ​​of the attention weight matrix, respectively.

6. The prism lens combination optical kit according to claim 5, characterized in that: If the eigenvalues ​​of the attention weight matrix are all 0 or there are no real eigenvalues, then λ = 0 and ξ = 1.

7. The prism lens combination optical kit according to claim 5, characterized in that: The method of obtaining a tight bonding index based on the contact area and the contact uniformity index includes: Obtain the projection area of ​​the first-dimensional contact vector mapped into the optical path image; The tight bonding index is obtained based on the contact uniformity index and the ratio K of the projected area to the contact area.

8. The prism lens combination optical kit according to claim 7, characterized in that: Based on the contact uniformity index and the ratio K of the projected area to the contact area, the tight bonding index is obtained, including: The contact uniformity index and the ratio K of the projected area to the contact area are weighted summed to obtain the tight bonding index, which is specifically obtained by the following formula: D=ξλU+(1-λξ)K Where D is the tight bonding index.

9. The prism lens combination optical kit according to claim 1, wherein: The prism lens combination optical kit also Includes the following components: Mounting part: a structure used to fix the optical kit; Kit structure: a frame for providing support and connection, with a card slot provided on one side of the kit structure; Prism lens combination optical element: the prism lens combination optical element is an optical element of an integrally formed prism lens combination; The prism-lens combination optical element is glued into the card slot of the kit structure, so that the prism-lens combination optical kit can realize the functions of prism and lens; the kit structure is connected to the mounting part on the other side away from the card slot; the mounting part is used to install the kit structure and the prism-lens combination optical element on the equipment; the installation detection component is arranged at a position of the kit structure away from the mounting part.

10. A method for producing a prism lens combination optical kit, characterized in that: The application is used for the process of producing the prism-lens combination optical kit, which includes a production and installation detection component, a mounting portion, a kit structure and a prism-lens combination optical element; Mounting part: a structure used to fix the optical kit; Kit structure: a frame for providing support and connection, with a card slot provided on one side of the kit structure; Prism lens combination optical element: the prism lens combination optical element is an optical element of an integrally formed prism lens combination; The prism-lens combination optical element is glued into the card slot of the kit structure, so that the prism-lens combination optical kit can realize the functions of a prism and a lens; the kit structure is connected to the mounting portion on the other side away from the card slot; the mounting portion is used to mount the kit structure and the prism-lens combination optical element on the device; the installation detection component is arranged at a position of the kit structure away from the mounting portion; Production and installation of detection components, including processors, cameras, and optical path detection components; A method for producing a prism lens combination optical kit, comprising: Capturing a cross-sectional image of a contact surface between the prism-lens combination optical element and the kit structure through a camera, and sending the cross-sectional image to a processor; obtaining optical path data of the prism-lens combination optical element when it is installed in the card slot through an optical path detection component, and sending the optical path data to the processor; The processor determines whether the prism lens combination optical element is tightly glued to the card slot based on the cross-sectional image and the optical path data, and determines whether the prism lens effect of the optical kit meets the standard; if not, a prompt message is issued; Wherein, judging whether the prism lens combination optical element is tightly glued to the card slot according to the cross-sectional image and the optical path data includes: Preprocessing: Denoise and enhance the cross-sectional image to obtain a preprocessed image; clean the optical path data to obtain cleaned optical path data; Extracting contact surface contours based on preprocessed images; The contact area and contact uniformity index are obtained based on the contact surface profile and the cleaning optical path data; the contact uniformity index is used to characterize the degree of contact uniformity between the prism lens combination optical element and the card slot; Based on the contact area and the contact uniformity index, a tight bonding index is obtained; the tight bonding index is used to indicate whether the prism lens combination optical element and the card slot are tightly bonded; If the tight bonding index is less than the set value, it is determined that the prism lens combination optical element is not tightly bonded to the card slot; If the tight bonding index is greater than or equal to the set value, it is determined that the prism lens combination optical element is tightly bonded to the card slot.

Images