Prism-lens combined optical suite and production method thereof

Abstract

The invention discloses a prism and lens combined optical suite and a production method thereof.The optical suite comprises an installation part, a suite structure, a prism and lens combined optical element and a production and installation detection component, the prism and lens combined optical element is glued in a clamping groove of the suite structure, and therefore the optical suite can achieve the functions of a prism and a lens; in the assembling production process, whether the prism and lens combined optical element is tightly glued to the clamping groove or not is detected, light path experiment data are collected, and whether the prism and lens effect of the optical suite reaches the standard or not is determined by detecting whether the prism and lens combined optical element is tightly glued to the clamping groove or not; if not, prompt information is sent out, the clamping groove and the position relation between the clamping groove and the prism-lens combined optical element are adjusted in time, the design and the production process are optimized, and the performance and the 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 optical instruments. Prisms and lenses are two basic optical components, each with unique functions and application areas.

[0003] Application and function of prism: prism can change the direction of light propagation and deflect the incident light through the principle of refraction. They are widely used in spectral analysis, imaging systems and optical instruments, and can achieve functions such as dispersion, reflection and light deflection. In cameras and projectors, prisms can be used to correct the direction of the image to ensure the accuracy and clarity of the image.

[0004] Characteristics and applications of lenses: Lenses are mainly used to focus or diverge light and can change the focal length of light. Convex lenses are used to focus light and are usually used for magnification and imaging, while concave lenses are used to diffuse light and are widely used in glasses and optical instruments. The design and manufacturing accuracy of lenses is crucial to their optical performance, especially in high-end optical equipment, where any slight error may 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, which leads to unstable optical paths and affects optical performance.

[0006] During the production process, although there are certain detection methods, there is still a lack of real-time monitoring and analysis of the bonding effect of optical components, which easily leads to fluctuations in product quality. In addition, traditional optical components are usually designed and manufactured independently, and cannot fully utilize the combined advantages of prisms and lenses, resulting in increased complexity of the optical system and increased 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 and improve production efficiency and product reliability.

[0009] In summary, the prism and lens optical components in the prior art have obvious deficiencies in functional integration, assembly accuracy and quality control. In view of these problems, the present invention proposes a novel prism lens combination optical kit and a production method thereof, 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 collect 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 on the card slot, and send the light path data to the processor;

[0013] The processor is used to determine 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, 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: De-noising and enhancing the cross-sectional image to obtain a preprocessed image; cleaning the optical path data to obtain cleaned optical path data;

[0016] Extracting contact surface contour 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 characterize whether the prism lens combination optical element is tightly bonded to the card slot;

[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 optical path data includes:

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

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

[0024] Extract the light intensity information of the contact area from the cleaning optical path data to obtain the first-dimensional contact vector; the first-dimensional contact vector represents the contact situation between the mirror assembly optical element and the card slot in the first dimension, and the first-dimensional contact vector is represented by the 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 combining the first dimension contact vector and the second dimension 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 dimension contact vector and the second dimension contact vector to obtain a contact uniformity index, including:

[0030] 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; 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 dimension contact vector and the second dimension contact vector.

[0031] Optionally, 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:

[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 A; The second branch hidden layer of the i-th layer extracts features from the second input of the i-th layer to obtain 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 ) indicates that the mapping function is for the input (W A H A +b A )’s mapping output, f(W B H B +b B ) indicates that the mapping function is for the input (W B H B +b B )’s mapping output;

[0037] Step S3: After the interaction layer, the fusion mechanism is used to combine the first mapping feature of the i-th layer and the second mapping feature of the i-th layer 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] Among them, 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, It means that the softmax function is 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 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 in the manner of steps S1 to S5;

[0043] The method shown in step S1 to step 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 adjustment feature and the last second branch adjustment feature;

[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 adjusted features of the first branch of the next layer, I A represents the first branch adjustment feature 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-fine number, ε = 0.001; σ(I B ) represents the standard deviation of the adjusted features of the first branch of the next layer, I B represents the first branch adjustment feature 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 dimension contact vector mapped to the optical path image;

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

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

[0051] The contact uniformity index and the ratio K of the projection 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 a prism lens combination formed in one piece;

[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 other side of the kit structure away from the card slot is connected to the mounting part; 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] The embodiment of the present invention also 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, wherein the prism-lens combination optical kit includes a production and installation detection component, an installation part, 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 a prism lens combination formed in one piece;

[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 prism and lens; the other side of the kit structure away from the card slot is connected to the mounting part; the mounting part 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 part;

[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] A cross-sectional image of a contact surface between the prism lens combination optical element and the kit structure is collected by a camera, and the cross-sectional image is sent to a processor; optical path data when the prism lens combination optical element is installed on the card slot is obtained by an optical path detection component, and the optical path data is sent to the processor;

[0067] The processor determines 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, 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: De-noising and enhancing the cross-sectional image to obtain a preprocessed image; cleaning the optical path data to obtain cleaned optical path data;

[0070] Extracting contact surface contour 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 characterize whether the prism lens combination optical element is tightly bonded to the card slot;

[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] The embodiment of the present invention provides a prism lens combination optical kit and a production method thereof, wherein a camera is used to collect a cross-sectional image of the contact surface between the prism lens combination optical element and the kit structure, the cross-sectional image is sent to a processor, and the optical path data of the prism lens combination optical element when it is installed on a card slot is obtained through an optical path detection component, 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 standard; if not, a prompt message is issued. Wherein, judging 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 contour based on the preprocessed image; obtain the contact area and contact uniformity index based on the contact surface contour 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 judging whether the optical element of the mirror assembly is tightly glued to the card slot can be improved, so as to improve the accuracy of determining whether the prism lens effect of the optical kit meets the standard. On this basis, the design and production process can be optimized to improve the performance and stability of the optical element. 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] Simplify the production process: By checking whether the optical components are tightly glued to the card slot during the production process, problems can be discovered and corrected in a timely manner, simplifying the production process and improving production efficiency. Data-driven quality control: Collecting and analyzing optical path experimental data can provide a basis for the quality control of optical kits, ensuring that each product meets the design standards and improving product consistency and reliability.

[0082] Adjustability and modularity: By adjusting the position 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] Strongly 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 according to an embodiment of the present invention. DETAILED DESCRIPTION

[0089] 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 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 and improving production efficiency and product reliability.

[0090] The present invention will be described in detail below in conjunction with 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, and the production and installation detection component includes a processor, a camera and an optical path detection component.

[0093] Among them, the camera is used to collect 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 the 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 standard; 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 standard.

[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: De-noising and enhancing the cross-sectional image to obtain a preprocessed image; cleaning the optical path data to obtain cleaned optical path data.

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

[0097] During preprocessing, the collected images are subjected to denoising, contrast enhancement and other preprocessing steps to improve the accuracy of subsequent analysis. Denoising can be performed using mean filtering, median filtering, Gaussian filtering and other methods, and enhancement can be performed using histogram equalization, brightness and contrast adjustment, Laplace filter and high-pass filter to enhance the edges and details of the image.

[0098] Cleaning optical path data: Record the light intensity distribution of the contact area and measure it using an optical sensor or camera. Data integration: Ensure that the optical path data is spatially aligned with the image data, which requires coordinate transformation or interpolation processing.

[0099] After preprocessing, the contact surface contour is extracted based on the preprocessed image. In the embodiment of the present invention, the contact surface contour can be extracted using a canny operator. The contour of the contact area can also be extracted from the processed image using an edge detection algorithm, and the contour is further cleaned using morphological operations (such as dilation and erosion) 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 the cleaning optical path data, wherein the contact uniformity index is used to characterize the degree of contact uniformity between the prism lens combination 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] 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. Constructing a light path image based on the light path data specifically includes using the light intensity of the light path data as the pixel value of the light path image, and using the light path illumination point as the coordinate point of the light path image to form the light path image.

[0104] The contact area is obtained based on the contact area. Specifically, the contact area can be detected using the Canny operator or the Huffman operator, and the contact area is binarized. The white area represents the contact area, and the black area represents the non-contact area. Calculate the number of white pixels in the binary image: multiply the number of white pixels by the area occupied by each pixel to get the contact area.

[0105] The 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 characterizes the contact between the mirror combination 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 the contact area is subjected to feature extraction to obtain a first-dimensional contact vector. The feature extraction method can use a convolutional neural network for feature extraction, or a principal component analysis method for principal component analysis to extract features and obtain a first-dimensional contact vector. In an embodiment of the present invention, the first dimension is information in the X direction of the coordinate axis, or dimensional information of a plane passing through the X axis, and the second dimension is information in the Y direction of the coordinate axis, or dimensional information on a plane perpendicular to the plane where the first dimension is located through the Y axis plane. That is, the first-dimensional contact vector is the feature information observed in the X direction, that is, the front of the contact surface is observed, and the second-dimensional contact vector is the feature information observed in the Y direction, that is, the cross section of the contact surface is observed.

[0106] The second-dimensional contact vector is obtained based on the contact surface profile. The second-dimensional contact vector characterizes the contact situation between the optical element of the mirror assembly and the card slot in the second dimension. That is, the second-dimensional contact vector is obtained by extracting features 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-dimensional contact vector and the second-dimensional contact vector, the contact uniformity index is obtained. Specifically:

[0108] 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. The layer perceptron model includes a first branch, a second branch and an interaction layer, the first branch includes multiple first branch hidden layers connected in sequence, and the second branch includes multiple second branch hidden layers connected in sequence; the interaction layer is set 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 the 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 dimension contact vector and the second dimension contact vector.

[0109] In an embodiment of the present invention, a combined analysis is performed on the first dimension contact vector and the second dimension contact vector by combining a multi-layer perceptron model, and the multi-layer perceptron model is combined to output 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 to obtain 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. 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 ) indicates that the mapping function is for the input (W A H A +b A )’s mapping output, f(W B H B +b B ) indicates that the mapping function is for the input (W B H B +b B )’s mapping output.

[0115] Step S3: After the interaction layer, the fusion mechanism is used to combine the first mapping feature of the i-th layer and the second mapping feature of the i-th layer 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, obtained by training the joint multi-layer perceptron model; A = Represents H′ A The transposed matrix of , softmax() is the softmax function, It means that the softmax function is 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 at 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 features of the i+1th layer and the second branch adjustment features of the i+1th 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 adjustment features of the i+1th layer and the second branch adjustment features of the i+1th layer are obtained in accordance with steps S1 to S5.

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

[0122] Step S7: Based on the last layer of first branch adjustment features and the last layer of second branch adjustment features, a contact uniformity index is obtained through the following formula.

[0123] Where U represents the contact uniformity index, σ(I A ) represents the standard deviation of the adjusted features of the first branch of the next layer, I A represents the first branch adjustment feature 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-fine number, ε = 0.001; σ(I B ) represents the standard deviation of the adjusted features of the first branch of the next layer, I B represents the first branch adjustment feature 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 model (Multi-Layer Perceptron, MLP) structure can effectively capture the complex correlation between them when processing two sets of data with different dimensions, directions and meanings through multiple input branches, feature fusion layers and attention mechanisms. This design not only improves the expressiveness of the model, but also enhances the flexibility and interpretability of the model. At the same time, by introducing the interaction layer and fusion mechanism, the first branch and the second branch can effectively share and fuse each other's implicit information, thereby enhancing the model's learning ability for complex data relationships, and can spread the implicit correlation between the first dimension contact vector and the second dimension contact vector. At the same time, the contact uniformity index obtained by the above method can accurately, quickly and effectively characterize the uniformity of the bonding between the mirror combination optical element and the card slot, thereby improving the accuracy and reliability of judging whether the mirror combination 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 is tightly bonded to the card slot.

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

[0128] The projection area of ​​the first-dimensional contact vector mapped to the optical path image is obtained. Based on the contact uniformity index and the ratio K of the projection area to the contact area, a tight bonding index is obtained. Specifically, the contact uniformity index and the ratio K of the projection area to the contact area are weightedly summed to obtain the tight bonding index, which is specifically obtained by 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 characterize the bonding degree between the prism lens combination optical element and the card slot.

[0132] As a further step, if the tight bonding index is less than a 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. In an embodiment of the present invention, the set value is between 0.05-20.

[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] Kit structure: 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 a prism lens combination formed in one piece.

[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 other side of the kit structure away from the card slot is connected to the mounting part; the mounting part is used to install the kit structure and the prism-lens combination optical original 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 when the prism lens combination optical element 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 according to 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 scheme, the prism and lens optical components in the prior art have obvious deficiencies in functional integration, assembly accuracy and quality control. In response to these problems, the present invention proposes a prism lens combination optical kit and a production method thereof, 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 element 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 such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the description of the above specific languages ​​is for disclosing the best mode of the present invention.

[0145] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can 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 present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the intention that the claimed invention requires more features than those explicitly recited in each claim. More specifically, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present 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 they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, 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] In addition, those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in the claims below, any one of the claimed embodiments may be used in any combination.

[0149] The various component embodiments of the present invention may 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) may 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 may also be implemented as a device or device program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing the present invention may be stored on a computer-readable medium, or may have the form of one or more signals. Such a signal may 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 the present invention rather than limit it, 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 symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" 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 a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. 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, the production and installation of detection components include processors, cameras and optical path detection components; The camera is used to collect 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 on the card slot, and send the light 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 according to 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: De-noising and enhancing the cross-sectional image to obtain a preprocessed image; cleaning the optical path data to obtain cleaned optical path data; Extracting contact surface contour 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 characterize whether the prism lens combination optical element is tightly bonded to the card slot; 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 step of obtaining the contact area and the contact uniformity index based on the contact surface profile and the cleaning optical path data includes: Constructing a light path image based on the light path data; extracting a contact area between the prism lens combination optical element and the card slot based on the light path image; Obtaining a contact area based on the contact region; Extract the light intensity information of the contact area from the cleaning optical path data to obtain the first-dimensional contact vector; the first-dimensional contact vector represents the contact situation between the mirror assembly optical element and the card slot in the first dimension, and the first-dimensional contact vector is represented by the 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 combining the first dimension contact vector and the second dimension contact vector.

3. The prism lens combination optical kit according to claim 2, characterized in that: The step of constructing a 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 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; 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 dimension contact vector and the second dimension 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 to obtain 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 ) indicates that the mapping function is for the input (W A H A +b A )’s mapping output, f(W B H B +b B ) indicates that the mapping function is for the input (W B H B +b B )’s mapping output; Step S3: After the interaction layer, the fusion mechanism is used to combine the first mapping feature of the i-th layer and the second mapping feature of the i-th layer 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 Among them, 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, It means that the softmax function is 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 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 in the manner of steps S1 to S5; The method shown in step S1 to step 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 adjustment feature and the last second branch adjustment feature; 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 adjusted features of the first branch of the next layer, I A represents the first branch adjustment feature 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-fine number, ε = 0.001; σ(I B ) represents the standard deviation of the adjusted features of the first branch of the next layer, I B represents the first branch adjustment feature 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 dimension contact vector mapped to the optical path image; Based on the contact uniformity index and the ratio K of the projected area to the contact area, the tight bonding index is obtained.

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 projection area to the contact area, the tight bonding index is obtained, including: The contact uniformity index and the ratio K of the projection 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, characterized in that: 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 a prism lens combination formed in one piece; 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 other side of the kit structure away from the card slot is connected to the mounting part; 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 in the process of producing the prism-lens combination optical kit, wherein 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; 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 a prism lens combination formed in one piece; 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 other side of the kit structure away from the card slot is connected to the mounting part; the mounting part 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 part; 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: A cross-sectional image of a contact surface between the prism lens combination optical element and the kit structure is collected by a camera, and the cross-sectional image is sent to a processor; optical path data when the prism lens combination optical element is installed on the card slot is obtained by an optical path detection component, 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 according to 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: De-noising and enhancing the cross-sectional image to obtain a preprocessed image; cleaning the optical path data to obtain cleaned optical path data; Extracting contact surface contour 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 characterize whether the prism lens combination optical element is tightly bonded to the card slot; 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