Image measuring instrument AI automatic point acquisition optimization processing method, system and platform based on adaptive learning

Abstract

The invention discloses an image measuring instrument AI automatic point acquisition optimization processing method, system and platform based on adaptive learning. The method comprises the following steps: generating and acquiring first data corresponding to an image measuring instrument to be optimized, and constructing and generating corresponding second data; extracting and processing the second data by adopting a CSP backbone network structure, and generating corresponding third data; according to a self-adaptive learning algorithm, anchor frame size data and loss function weight data of the model are dynamically adjusted respectively, and fourth data are created and generated in combination with multiple data enhancement technologies; based on the fourth data, creating and generating a first model, marking and processing the first data according to the first model, and generating fifth data; constructing a second model according to the fifth data, and generating corresponding sixth data; and the sixth data is measurement data corresponding to the product to be measured, and the system and the platform corresponding to the method dynamically optimize the AI point acquisition process through an adaptive learning algorithm, thereby improving the measurement efficiency and precision.

Description

Technical Field

[0001] The present invention belongs to the technical field of image measurement and processing, and specifically relates to an image measuring instrument AI automatic point sampling optimization processing method, system and platform based on adaptive learning. Background Art

[0002] In the dimensional measurement of industrial products, image measuring instruments are widely used due to their non-contact and high-precision characteristics. However, in the manual mouse point measurement of existing image measuring instruments, users are required to click on the measurement points one by one, which is cumbersome and time-consuming, and it is difficult to meet the needs of high-efficiency production; moreover, the accuracy of manual point measurement is highly dependent on the operator's experience and technical level, which is easy to introduce human errors and lead to unstable measurement results; in addition, existing automatic measurement methods usually use fixed parameters and rules, and cannot dynamically adjust the point measurement strategy according to the characteristics of the measurement object, resulting in poor adaptability to complex surface features.

[0003] Although some image measuring instruments have introduced automatic measurement functions, they usually rely on preset geometric models or rules and are difficult to cope with the complex and changing characteristics of industrial products. In addition, existing automatic measurement methods lack the ability to adaptively learn sample data and cannot dynamically optimize model parameters according to changes in measurement tasks, making it difficult to further improve measurement accuracy and efficiency.

[0004] Therefore, in view of the above-mentioned technical problems such as manual point sampling efficiency bottleneck, accuracy dependence on manual experience and lack of adaptive capabilities, it is urgent to design and develop an image measuring instrument AI automatic point sampling optimization processing method, system and platform based on adaptive learning. Summary of the invention

[0005] In order to overcome the shortcomings and difficulties of the above-mentioned prior art, the present invention provides an image measuring instrument AI automatic point sampling optimization processing method, system and platform based on adaptive learning. Through the adaptive learning algorithm, the process of AI automatic point sampling is optimized to achieve efficient and high-precision automatic measurement.

[0006] The first object of the present invention is to provide an image measuring instrument AI automatic point sampling optimization processing method based on adaptive learning; the second object of the present invention is to provide an image measuring instrument AI automatic point sampling optimization processing system based on adaptive learning; the third object of the present invention is to provide an image measuring instrument AI automatic point sampling optimization processing platform based on adaptive learning; The first object of the present invention is achieved in that the method comprises the following steps: Generate and obtain first data corresponding to the image measuring instrument to be optimized, and construct and generate corresponding second data based on the first data; wherein the first data is the image area measurement point data; and the second data is the sample data that has been marked; Adopting the CSP backbone network structure, extracting and processing the second data, and generating corresponding third data; wherein the third data is characteristic data of the sample data; According to the adaptive learning algorithm, the anchor box size data and the loss function weight data of the model are dynamically adjusted respectively, and a plurality of data enhancement techniques are combined to create and generate corresponding fourth data; wherein the fourth data is sample data with diversity; Based on the fourth data, a corresponding first model is created and generated, and according to the first model, the first data is marked and processed to generate corresponding fifth data; wherein the first model is a deep learning model for automatic point sampling; and the fifth data is point sampling location information data; According to the fifth data, a second model corresponding to the measurement is constructed, and corresponding sixth data is generated based on the second model; wherein the second model is a measurement element model; and the sixth data is measurement data corresponding to the product to be measured.

[0007] Furthermore, the method of generating and acquiring first data corresponding to the image measuring instrument to be optimized, and constructing and generating corresponding second data based on the first data, also includes: Generate and obtain seventh data corresponding to the product to be measured and processed; wherein the seventh data is activation control data for starting to collect product image data; According to the seventh data, the first data is marked and processed, and the second data is recorded and maintained.

[0008] Furthermore, the method of adopting the CSP backbone network structure to extract and process the second data and generate corresponding third data also includes: Based on the CSP backbone network structure, third data corresponding to the second data is calculated and generated; wherein the calculation formula is as follows: (1) In the formula, is the fused feature; Further, the method dynamically adjusts the anchor box size data and loss function weight data of the model according to the adaptive learning algorithm, and combines a variety of data enhancement techniques to create and generate corresponding fourth data, and also includes: Combined with the adaptive learning algorithm, the anchor box size and loss function weight of the model are dynamically adjusted; the calculation formula of the anchor box size is as follows: (2) In the formula, For the Anchor box size; is the prediction box; is the real frame; for intersection and comparison; Based on data enhancement technology, the second data is processed in real time to generate corresponding eighth data; wherein the processing includes random scaling processing, cropping processing, flipping processing and translation processing; the eighth data is diversified sample data corresponding to the sample data.

[0009] Further, the creating and generating a corresponding first model based on the fourth data, and marking and processing the first data according to the first model to generate corresponding fifth data, further includes: Based on the first model and in combination with the first data, corresponding fifth data is calculated and generated; wherein the calculation formula is as follows: (3) In the formula, For automatic sampling point location; For location confidence level; For location With real frame The intersection and union ratio of is the area range of the points to be sampled in the image. Further, the second model corresponding to the measurement is constructed according to the fifth data, and the corresponding sixth data is generated based on the second model, and further includes: According to the fifth data, construct at least one corresponding measurement element; wherein the measurement element includes a line, an arc and a distance; Combined with stochastic gradient descent or Adam optimizer, the convolutional neural network-based object detection model or key point detection model is trained and optimized respectively.

[0010] The second object of the present invention is achieved as follows: the system is used to implement the image measuring instrument AI automatic point sampling optimization processing method based on adaptive learning, and the system includes: A first data generating unit is used to generate and obtain first data corresponding to the image measuring instrument to be optimized, and to construct and generate corresponding second data based on the first data; wherein the first data is the image area measurement point data; and the second data is the sample data that has been marked; A second data generating unit, configured to extract and process the second data by adopting a CSP backbone network structure, and generate corresponding third data; wherein the third data is characteristic data of the sample data; A first data processing unit is used to dynamically adjust the anchor box size data and loss function weight data of the model according to the adaptive learning algorithm, and combine multiple data enhancement technologies to create and generate corresponding fourth data; wherein the fourth data is sample data with diversity; A second data processing unit is configured to create and generate a corresponding first model based on the fourth data, and to mark and process the first data according to the first model to generate corresponding fifth data; wherein the first model is a deep learning model for automatic point sampling; and the fifth data is point sampling location information data; The third data generating unit is used to construct a second model corresponding to the measurement according to the fifth data, and generate corresponding sixth data based on the second model; wherein the second model is a measurement element model; and the sixth data is measurement data corresponding to the product to be measured.

[0011] Furthermore, the first data generating unit further includes: A first generating module is used to generate and obtain seventh data corresponding to the product to be measured and processed; wherein the seventh data is activation control data for starting to collect product image data; A first processing module, used for marking and processing the first data according to the seventh data, and recording and maintaining the second data; And / or, the second data generating unit further includes: The first calculation module is used to calculate and generate third data corresponding to the second data based on the CSP backbone network structure; wherein the calculation formula is as follows: (1) In the formula, is the fused feature; are feature maps at different stages; and / or, the first data processing unit further includes: The second processing module is used to dynamically adjust the anchor box size and loss function weight of the model in combination with the adaptive learning algorithm; wherein the anchor box size is calculated as follows: (2) In the formula, For the Anchor box size; is the prediction box; is the real frame; is an intersection-and-union ratio; a second generating module is used to process the second data in real time based on data enhancement technology, and generate corresponding eighth data; wherein the processing includes random scaling processing, cropping processing, flipping processing and translation processing; the eighth data is diversified sample data corresponding to the sample data; And / or, the second data processing unit further includes: The second calculation module is used to calculate and generate corresponding fifth data based on the first model and in combination with the first data; wherein the calculation formula is as follows: (3) In the formula, For automatic sampling point location; For location confidence level; For location With real box The intersection and union ratio of is the area range of the image to be sampled; and / or, the third data generating unit further includes: A first construction module, configured to construct at least one corresponding measurement element according to the fifth data; wherein the measurement element includes a line, an arc and a distance; The third processing module is used to combine the stochastic gradient descent or Adam optimizer to respectively train and optimize the target detection model or the key point detection model based on the convolutional neural network.

[0012] The third object of the present invention is achieved as follows: it includes a processor, a memory, and an image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning; wherein the image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning is executed in the processor, and the image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning is stored in the memory, and the image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning implements the image measuring instrument AI automatic point sampling optimization processing method based on adaptive learning.

[0013] The present invention generates and obtains first data corresponding to the image measuring instrument to be optimized through a method, and constructs and generates corresponding second data based on the first data; wherein the first data is the image area measurement point data; the second data is the sample data that has been marked; the CSP backbone network structure is adopted to extract and process the second data, and generate corresponding third data; wherein the third data is the feature data of the sample data; according to the adaptive learning algorithm, the anchor frame size data and the loss function weight data of the model are dynamically adjusted respectively, and a plurality of data enhancement technologies are combined to create and generate corresponding fourth data; wherein the fourth data is the sample data with diversity; based on the fourth data, a corresponding first model is created and generated, and according to the first model, the first data is marked and processed to generate corresponding fifth data; wherein the first model is a deep learning model for automatic point sampling; the fifth data is point sampling location information data; according to the fifth data, a second model corresponding to the measurement is constructed, and corresponding sixth data is generated based on the second model; wherein the second model is a measurement element model; the sixth data is the measurement data corresponding to the product to be measured, and a system and platform corresponding to the method dynamically optimize the AI ​​point sampling process through an adaptive learning algorithm to improve measurement efficiency and accuracy.

[0014] That is to say, the scheme of the present invention optimizes the process of automatic point collection by AI by introducing an adaptive learning algorithm, thereby solving the problems of low efficiency, unstable accuracy and lack of adaptive ability in the prior art; through the adaptive learning algorithm, the anchor frame size and loss function weight of the model are dynamically adjusted to optimize the calculation of the point collection position and improve the measurement accuracy; the automatic point collection process is completely completed by the AI ​​model, which significantly reduces the measurement time, and is particularly suitable for large-scale measurement tasks, that is, improving the measurement efficiency; through data enhancement technology and adaptive learning algorithms, the model's adaptability to different measurement scenarios is improved, and it is suitable for the measurement of complex surface features; the automatic point collection process avoids errors introduced by human operations and ensures the consistency of measurement results; through the adaptive learning algorithm, the model can be continuously optimized according to new sample data to maintain high measurement accuracy and robustness; through lightweight design and optimizer selection, the model parameters and calculation amount are reduced, the running speed is improved, and it is suitable for real-time measurement scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0016] Figure 1 It is a schematic diagram of the process steps of an AI automatic point sampling optimization processing method of an image measuring instrument based on adaptive learning of the present invention; Figure 2 A schematic diagram of the conversion process steps of the manual mouse point collection measurement into AI automatic point collection of an image measuring instrument AI automatic point collection optimization processing method based on adaptive learning of the present invention; Figure 3 A schematic diagram of the steps of the manual mouse point collection measurement process without using AI automatic point collection in an image measuring instrument AI automatic point collection optimization processing method based on adaptive learning of the present invention; Figure 4 It is a schematic diagram of the measurement process steps after manual mouse point collection is converted into AI automatic point collection in an image measuring instrument AI automatic point collection optimization processing method based on adaptive learning of the present invention; Figure 5 This is a schematic diagram of the architecture of an AI automatic point sampling optimization processing system of an image measuring instrument based on adaptive learning in the present invention; Figure 6 This is a schematic diagram of the architecture of an AI automatic point sampling optimization processing platform for an image measuring instrument based on adaptive learning in the present invention. DETAILED DESCRIPTION

[0017] In order to better understand the purpose, technical solutions and advantages of the present invention, the present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification.

[0018] The present invention may also be implemented or applied through other different specific examples, and the details in this specification may also be modified and changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0019] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0020] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. Secondly, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0021] Preferably, the AI ​​automatic point sampling optimization processing method of the image measuring instrument based on adaptive learning is applied in one or more terminals or servers. The terminal is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), digital signal processors (DSP), embedded devices, etc.

[0022] The terminal can be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal can interact with the client through a keyboard, a mouse, a remote control, a touch pad, or a voice control device.

[0023] The present invention is to realize an image measuring instrument AI automatic point sampling optimization processing method, system and platform based on adaptive learning.

[0024] like Figure 1 , which is a flow chart of an image measuring instrument AI automatic point sampling optimization processing method based on adaptive learning provided by an embodiment of the present invention.

[0025] In this embodiment, the image measuring instrument AI automatic point sampling optimization processing method based on adaptive learning can be applied to a terminal or a fixed terminal with a display function. The terminal is not limited to a personal computer, a smart phone, a tablet computer, a desktop computer or an all-in-one computer equipped with a camera, etc.

[0026] The method for optimizing the automatic point sampling of an image measuring instrument based on adaptive learning can also be applied to a hardware environment consisting of a terminal and a server connected to the terminal through a network. The network includes but is not limited to: a wide area network, a metropolitan area network or a local area network. The method for optimizing the automatic point sampling of an image measuring instrument based on adaptive learning in an embodiment of the present invention can be executed by a server, by a terminal, or by both a server and a terminal.

[0027] For example, for a terminal that needs to perform adaptive learning-based image measuring instrument AI automatic point sampling optimization processing, the adaptive learning-based image measuring instrument AI automatic point sampling optimization processing function provided by the method of the present invention can be directly integrated on the terminal, or a client for implementing the method of the present invention can be installed. For another example, the method provided by the present invention can also be run on a server or other device in the form of a software development kit (SDK), and an interface for the adaptive learning-based image measuring instrument AI automatic point sampling optimization processing function is provided in the form of SDK. The terminal or other device can implement the adaptive learning-based image measuring instrument AI automatic point sampling optimization processing function through the provided interface. The present invention is further described below in conjunction with the accompanying drawings.

[0028] like Figure 1 As shown, the present invention provides an AI automatic point sampling optimization processing method for an image measuring instrument based on adaptive learning, wherein the method comprises the following steps: S1. Generate and obtain first data corresponding to the image measuring instrument to be optimized, and construct and generate corresponding second data based on the first data; wherein the first data is the image area measurement point data; and the second data is the sample data that has been marked; S2. Using the CSP backbone network structure, extract and process the second data, and generate corresponding third data; wherein the third data is feature data of the sample data; S3. According to the adaptive learning algorithm, dynamically adjust the anchor box size data and loss function weight data of the model respectively, and combine multiple data enhancement techniques to create and generate corresponding fourth data; wherein the fourth data is sample data with diversity; S4. Based on the fourth data, create a corresponding first model, and mark and process the first data according to the first model to generate corresponding fifth data; wherein the first model is a deep learning model for automatic point sampling; and the fifth data is point sampling location information data; S5. Construct a second model corresponding to the measurement according to the fifth data, and generate corresponding sixth data based on the second model; wherein the second model is a measurement element model; and the sixth data is measurement data corresponding to the product to be measured.

[0029] The generating and acquiring first data corresponding to the image measuring instrument to be optimized, and constructing and generating corresponding second data based on the first data; further comprising: S11, generating and acquiring seventh data corresponding to the product to be measured and processed; wherein the seventh data is activation control data for starting to collect product image data; S12. Mark and process the first data according to the seventh data, and record and maintain the second data.

[0030] The method of adopting the CSP backbone network structure to extract and process the second data and generate corresponding third data also includes: S21. Based on the CSP backbone network structure, calculate and generate third data corresponding to the second data; wherein the calculation formula is as follows: (1) In the formula, is the fused feature; The method of dynamically adjusting the anchor box size data and loss function weight data of the model according to the adaptive learning algorithm, and combining a plurality of data enhancement techniques to create and generate corresponding fourth data, further includes: S31. Combined with the adaptive learning algorithm, dynamically adjust the anchor box size and loss function weight of the model; the calculation formula of the anchor box size is as follows: (2) In the formula, For the Anchor box size; is the prediction box; is the real frame; for intersection and comparison; S32. Based on data enhancement technology, the second data is processed in real time to generate corresponding eighth data; wherein the processing includes random scaling, cropping, flipping and translation; the eighth data is diversified sample data corresponding to the sample data.

[0031] The step of creating and generating a corresponding first model based on the fourth data, and marking and processing the first data according to the first model to generate corresponding fifth data further includes: S41. Based on the first model and in combination with the first data, calculate and generate corresponding fifth data; wherein the calculation formula is as follows: (3) In the formula, For automatic sampling point location; For location confidence level; For location With real frame The intersection and union ratio of The method of constructing a second model corresponding to the measurement based on the fifth data and generating corresponding sixth data based on the second model also includes: S51. Construct at least one corresponding measurement element according to the fifth data; wherein the measurement element includes a line, an arc and a distance; S52. Combined with stochastic gradient descent or Adam optimizer, respectively train and optimize the target detection model or key point detection model based on convolutional neural network.

[0032] Specifically, in an embodiment of the present invention, an image measuring instrument AI automatic point sampling optimization processing method based on adaptive learning is provided, which dynamically optimizes the AI ​​point sampling process through an adaptive learning algorithm to improve measurement efficiency and accuracy; the method comprises the following steps: the user opens the measurement program in the image measuring instrument software and selects the "adaptive learning AI automatic measurement" mode; places the product to be measured and starts the measurement program; the user manually operates the mouse to mark the measurement points in the image area, and the software automatically records these points as sample data and saves them in the database; the CSP (Cross Stage Partial) backbone network structure is used to extract the features of the sample data, and the feature extraction formula is: (1) In the formula, is the fused feature; are feature maps at different stages; the anchor box size and loss function weight of the model are dynamically adjusted through the adaptive learning algorithm. The anchor box size calculation formula is: (2) In the formula, For the Anchor box size; is the prediction box; is the real frame; is the intersection-and-union ratio. A variety of data enhancement techniques are used to increase the diversity of sample data; according to the trained model, the software automatically marks the measurement points in the image, and the calculation formula for the sampling point position is: (3) In the formula, For automatic sampling point location; For location confidence level; For location With real frame The intersection and union ratio of The area to be sampled in the image. Based on the location information of the automatically sampled points, the software constructs other measurement elements, including lines, arcs, and distances; outputs measurement data and reports to complete the measurement task; and uses stochastic gradient descent (SGD) or Adam optimizer to continuously train and optimize the model.

[0033] The adaptive learning algorithm optimizes the classification and positioning accuracy of the model by dynamically adjusting the anchor box size and loss function weight of the model. The data enhancement technology includes at least one of random scaling, cropping, flipping and translation. The CSP backbone network structure splits and fuses feature maps at different stages to enhance the propagation and reuse of features. The location information of the automatic point collection is recorded and saved in the database as reference data for subsequent measurements. The construction of other measurement elements includes at least one of lines, arcs and distances. The model reduces the number of parameters and calculations through lightweight design and improves the running speed. The optimizer includes stochastic gradient descent (SGD) and its variant Adam. The method continuously optimizes the accuracy and efficiency of automatic point collection through an adaptive learning algorithm. The method is suitable for the size measurement of industrial products with complex surface features.

[0034] Example: Implementation of adaptive learning and AI automatic point selection optimization For data collection and sample construction, the user opens the measurement program in the image measuring instrument software and selects the "Adaptive Learning AI Automatic Measurement" mode; places the product to be measured (such as electronic components) and starts the measurement program; the user manually operates the mouse to mark the measurement points in the image area (such as the pin position of the component), and the software automatically records these points as sample data and saves them in the database.

[0035] Adaptive learning and model training, using the CSP (Cross Stage Partial) backbone network structure to extract features of sample data. For example, the input image size is 512×512 pixels, and after passing through the CSP network, the output feature map size is 16×16×256; the anchor box size and loss function weight of the model are dynamically adjusted through the adaptive learning algorithm. For example, according to the distribution of sample data, the anchor box size is dynamically adjusted to [16, 32, 64] pixels; a variety of data enhancement techniques are used to increase the diversity of sample data. For example, the image is randomly rotated (rotation angle range is -30°~30°) and color jittered (brightness and contrast adjustment range is 0.8~1.2).

[0036] AI automatically collects points and optimizes them. Based on the trained model, the software automatically marks the measurement points in the image. For example, to mark the pin positions of components in the image, the calculation formula for the collected point positions is: (3) In the formula, For automatic sampling point location; For location confidence level; For location With real frame The intersection and union ratio of The area to be sampled in the image. Measurement and result output: The software constructs other measurement elements based on the position information of the automatically sampled points. For example, the pin spacing is calculated based on the marked pin position; the measurement data and report are output to complete the measurement task.

[0037] Model update and maintenance, using Adam optimizer, initial learning rate of 0.001, continuous training and optimization of the model; through lightweight design, reduce the model parameters and calculation amount. For example, compress the model parameters from 50MB to 5MB to improve the running speed.

[0038] The core of the solution of the present invention is to optimize the AI ​​automatic point collection process through an adaptive learning algorithm. Specifically, in adaptive learning, the model's classification and positioning accuracy are optimized by dynamically adjusting the model's anchor frame size and loss function weight; for AI automatic point collection, the location of the measurement point is automatically determined through confidence calculation and intersection over union (IOU) optimization to reduce human operation errors; in terms of model optimization, the model's running speed and measurement efficiency are improved through lightweight design and optimizer selection.

[0039] In another embodiment of the solution of the present invention, a method for converting the elements of mouse point measurement from manual measurement to AI automatic measurement (i.e., a method for converting manual point measurement to automatic point measurement) in a two-dimensional image measurement device is provided; Figure 2-Figure 4As shown in the figure, since manual work is replaced automatically, the measurement efficiency is greatly improved, and the measurement accuracy is also guaranteed through AI feature training (that is, the user's measurement efficiency is improved and the measurement accuracy is improved). Among them, the main technologies used in AI automatic point collection are: network structure design, using the CSP (Cross Stage Partial) backbone network structure to split and fuse feature maps at different stages, and enhance feature propagation and reuse.

[0040] Data enhancement technology uses a variety of data enhancement techniques, such as random scaling, cropping, flipping, translation, etc., to increase the diversity of data, enabling the model to learn more target features in different scenarios and improve the model's generalization ability and detection accuracy.

[0041] The training strategy is optimized and adaptive anchor box calculation is used. Before training, the anchor box size suitable for the data set is adaptively calculated through analysis of the data set, so that the model can better match the real bounding box of the target and improve the positioning accuracy.

[0042] The loss function design adopts a multi-part loss function including classification loss, regression loss and confidence loss to conduct comprehensive supervised learning on the different outputs of the model to ensure that the model can achieve high accuracy in classification and positioning.

[0043] Optimizer selection usually uses stochastic gradient descent (SGD) and its variants, such as Adam and other optimizers, and adjusts hyperparameters such as learning rate so that the model can quickly converge to the optimal solution, thereby ensuring detection accuracy.

[0044] Model optimization and improvement, under the premise of ensuring detection accuracy, the model is designed to be lightweight, the number of model parameters and the amount of calculation are reduced, and the running speed of the model is improved. It also helps to prevent overfitting and further improve detection accuracy.

[0045] To achieve the above purpose, the present invention also provides an image measuring instrument AI automatic point sampling optimization processing system based on adaptive learning, such as Figure 5 As shown, the system specifically includes: A first data generating unit is used to generate and obtain first data corresponding to the image measuring instrument to be optimized, and to construct and generate corresponding second data based on the first data; wherein the first data is the image area measurement point data; and the second data is the sample data that has been marked; A second data generating unit, configured to extract and process the second data by adopting a CSP backbone network structure, and generate corresponding third data; wherein the third data is characteristic data of the sample data; A first data processing unit is used to dynamically adjust the anchor box size data and loss function weight data of the model according to the adaptive learning algorithm, and combine multiple data enhancement technologies to create and generate corresponding fourth data; wherein the fourth data is sample data with diversity; A second data processing unit is configured to create and generate a corresponding first model based on the fourth data, and to mark and process the first data according to the first model to generate corresponding fifth data; wherein the first model is a deep learning model for automatic point sampling; and the fifth data is point sampling location information data; The third data generating unit is used to construct a second model corresponding to the measurement according to the fifth data, and generate corresponding sixth data based on the second model; wherein the second model is a measurement element model; and the sixth data is measurement data corresponding to the product to be measured.

[0046] The first data generating unit further includes: A first generating module is used to generate and obtain seventh data corresponding to the product to be measured and processed; wherein the seventh data is activation control data for starting to collect product image data; A first processing module, used for marking and processing the first data according to the seventh data, and recording and maintaining the second data; And / or, the second data generating unit further includes: The first calculation module is used to calculate and generate third data corresponding to the second data based on the CSP backbone network structure; wherein the calculation formula is as follows: (1) In the formula, is the fused feature; It is the feature map of different stages; And / or, the first data processing unit further includes: The second processing module is used to dynamically adjust the anchor box size and loss function weight of the model in combination with the adaptive learning algorithm; wherein the anchor box size is calculated as follows: (2) In the formula, For the Anchor box size; is the prediction box; is the real frame; for intersection and comparison; A second generating module is used to process the second data in real time based on data enhancement technology, and generate corresponding eighth data; wherein the processing includes random scaling processing, cropping processing, flipping processing and translation processing; the eighth data is diversified sample data corresponding to the sample data; And / or, the second data processing unit further includes: The second calculation module is used to calculate and generate corresponding fifth data based on the first model and in combination with the first data; wherein the calculation formula is as follows: (3) In the formula, For automatic sampling point location; For location Confidence level; For location With real box The intersection and union ratio of is the area to be sampled in the image; And / or, the third data generating unit further includes: A first construction module, configured to construct at least one corresponding measurement element according to the fifth data; wherein the measurement element includes a line, an arc and a distance; The third processing module is used to combine the stochastic gradient descent or Adam optimizer to respectively train and optimize the target detection model or the key point detection model based on the convolutional neural network.

[0047] In the system solution embodiment of the present invention, the method steps involved in the AI ​​automatic point sampling optimization processing of the image measuring instrument based on adaptive learning have been described above in detail, that is, the functional modules in the system are used to implement the steps or sub-steps in the above method embodiment, which will not be repeated here.

[0048] To achieve the above objectives, the present invention also provides an AI automatic point sampling optimization processing platform for image measuring instruments based on adaptive learning, such as Figure 6 As shown, it includes a processor, a memory, and an image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning; wherein, the image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning is executed in the processor, the image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning is stored in the memory, and the image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning implements the image measuring instrument AI automatic point sampling optimization processing method steps based on adaptive learning. For example: S1. Generate and obtain first data corresponding to the image measuring instrument to be optimized, and construct and generate corresponding second data based on the first data; wherein the first data is the image area measurement point data; and the second data is the sample data that has been marked; S2. Using the CSP backbone network structure, extract and process the second data, and generate corresponding third data; wherein the third data is feature data of the sample data; S3. According to the adaptive learning algorithm, dynamically adjust the anchor box size data and loss function weight data of the model respectively, and combine multiple data enhancement techniques to create and generate corresponding fourth data; wherein the fourth data is sample data with diversity; S4. Based on the fourth data, create a corresponding first model, and mark and process the first data according to the first model to generate corresponding fifth data; wherein the first model is a deep learning model for automatic point sampling; and the fifth data is point sampling location information data; S5. Construct a second model corresponding to the measurement according to the fifth data, and generate corresponding sixth data based on the second model; wherein the second model is a measurement element model; and the sixth data is measurement data corresponding to the product to be measured.

[0049] The specific details of the steps have been explained above and will not be repeated here.

[0050] In the embodiment of the present invention, the built-in processor of the AI ​​automatic point sampling optimization processing platform of the image measuring instrument based on adaptive learning can be composed of integrated circuits, for example, it can be composed of a single packaged integrated circuit, or it can be composed of multiple integrated circuits with the same function or different functions, including one or more central processing units (CPU), microprocessors, digital processing chips, graphics processors and various control chips. The processor uses various interfaces and lines to connect various components, and executes or executes programs or units stored in the memory, and calls data stored in the memory to perform various functions and process data of the AI ​​automatic point sampling optimization processing of the image measuring instrument based on adaptive learning; The memory is used to store program codes and various data. It is installed in the AI ​​automatic point sampling optimization processing platform of the image measuring instrument based on adaptive learning, and realizes high-speed and automatic access to programs or data during operation.

[0051] The memory includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electronically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0052] The present invention generates and obtains first data corresponding to the image measuring instrument to be optimized through a method, and constructs and generates corresponding second data based on the first data; wherein the first data is the image area measurement point data; the second data is the sample data that has been marked; the CSP backbone network structure is adopted to extract and process the second data, and generate corresponding third data; wherein the third data is the feature data of the sample data; according to the adaptive learning algorithm, the anchor frame size data and the loss function weight data of the model are dynamically adjusted respectively, and a plurality of data enhancement technologies are combined to create and generate corresponding fourth data; wherein the fourth data is the sample data with diversity; based on the fourth data, a corresponding first model is created and generated, and according to the first model, the first data is marked and processed to generate corresponding fifth data; wherein the first model is a deep learning model for automatic point sampling; the fifth data is point sampling location information data; according to the fifth data, a second model corresponding to the measurement is constructed, and corresponding sixth data is generated based on the second model; wherein the second model is a measurement element model; the sixth data is the measurement data corresponding to the product to be measured, and a system and platform corresponding to the method dynamically optimize the AI ​​point sampling process through an adaptive learning algorithm to improve measurement efficiency and accuracy.

[0053] That is to say, the scheme of the present invention optimizes the process of automatic point collection by AI by introducing an adaptive learning algorithm, thereby solving the problems of low efficiency, unstable accuracy and lack of adaptive ability in the prior art; through the adaptive learning algorithm, the anchor frame size and loss function weight of the model are dynamically adjusted to optimize the calculation of the point collection position and improve the measurement accuracy; the automatic point collection process is completely completed by the AI ​​model, which significantly reduces the measurement time, and is particularly suitable for large-scale measurement tasks, that is, improving the measurement efficiency; through data enhancement technology and adaptive learning algorithms, the model's adaptability to different measurement scenarios is improved, and it is suitable for the measurement of complex surface features; the automatic point collection process avoids errors introduced by human operations and ensures the consistency of measurement results; through the adaptive learning algorithm, the model can be continuously optimized according to new sample data to maintain high measurement accuracy and robustness; through lightweight design and optimizer selection, the model parameters and calculation amount are reduced, the running speed is improved, and it is suitable for real-time measurement scenarios.

[0054] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. An AI automatic point sampling optimization processing method for an image measuring instrument based on adaptive learning, characterized in that: The method comprises the steps of: Generate and obtain first data corresponding to the image measuring instrument to be optimized, and construct and generate corresponding second data based on the first data; wherein the first data is the image area measurement point data; the second data is the sample data that has been marked; Adopting the CSP backbone network structure, extracting and processing the second data, and generating corresponding third data; wherein the third data is characteristic data of the sample data; According to the adaptive learning algorithm, the anchor box size data and the loss function weight data of the model are dynamically adjusted respectively, and a plurality of data enhancement techniques are combined to create and generate corresponding fourth data; wherein the fourth data is sample data with diversity; Based on the fourth data, a corresponding first model is created and generated, and according to the first model, the first data is marked and processed to generate corresponding fifth data; wherein the first model is a deep learning model for automatic point sampling; and the fifth data is point sampling location information data; According to the fifth data, a second model corresponding to the measurement is constructed, and corresponding sixth data is generated based on the second model; wherein the second model is a measurement element model; and the sixth data is measurement data corresponding to the product to be measured.

2. According to the method of claim 1, the method is characterized in that: The generating and acquiring first data corresponding to the image measuring instrument to be optimized, and constructing and generating corresponding second data based on the first data; further comprising: Generate and obtain seventh data corresponding to the product to be measured and processed; wherein the seventh data is activation control data for starting to collect product image data; According to the seventh data, the first data is marked and processed, and the second data is recorded and maintained.

3. According to the method of claim 1, the method is characterized in that: The method of adopting the CSP backbone network structure to extract and process the second data and generate corresponding third data also includes: Based on the CSP backbone network structure, third data corresponding to the second data is calculated and generated; wherein the calculation formula is as follows: (1) In the formula, The fused features; The feature maps at different stages.

4. According to the method of claim 1, the method is characterized in that: The method of dynamically adjusting the anchor box size data and the loss function weight data of the model according to the adaptive learning algorithm, and combining a plurality of data enhancement techniques to create and generate corresponding fourth data, further includes: Combined with the adaptive learning algorithm, the anchor box size and loss function weight of the model are dynamically adjusted; the calculation formula of the anchor box size is as follows: (2) In the formula, For the Anchor box size; is the prediction box; is the real frame; is the intersection and comparison; Based on data enhancement technology, the second data is processed in real time to generate corresponding eighth data; wherein the processing includes random scaling processing, cropping processing, flipping processing and translation processing; the eighth data is diversified sample data corresponding to the sample data.

5. According to the method of claim 1, the method is characterized in that: The step of creating and generating a corresponding first model based on the fourth data, and marking and processing the first data according to the first model to generate corresponding fifth data further includes: Based on the first model and in combination with the first data, corresponding fifth data is calculated and generated; wherein the calculation formula is as follows: (3) In the formula, For automatic sampling point location; For location Confidence level; For location With real frame The intersection and union ratio of The area in the image where points are to be collected.

6. The method for automatic point sampling optimization processing of an image measuring instrument based on adaptive learning according to claim 1 is characterized in that: The method of constructing a second model corresponding to the measurement according to the fifth data, and generating corresponding sixth data based on the second model, further includes: According to the fifth data, construct at least one corresponding measurement element; wherein the measurement element includes a line, an arc and a distance; Combined with stochastic gradient descent or Adam optimizer, the convolutional neural network-based object detection model or key point detection model is trained and optimized respectively.

7. An AI automatic point sampling optimization processing system for image measuring instruments based on adaptive learning, characterized in that: The system is used to implement the AI ​​automatic point sampling optimization processing method of the image measuring instrument based on adaptive learning as described in any one of claims 1 to 6, and the system includes: A first data generating unit is used to generate and obtain first data corresponding to the image measuring instrument to be optimized, and to construct and generate corresponding second data based on the first data; wherein the first data is the image area measurement point data; and the second data is the sample data that has been marked; A second data generating unit, configured to extract and process the second data by adopting a CSP backbone network structure, and generate corresponding third data; wherein the third data is characteristic data of the sample data; A first data processing unit is used to dynamically adjust the anchor box size data and loss function weight data of the model according to the adaptive learning algorithm, and combine multiple data enhancement technologies to create and generate corresponding fourth data; wherein the fourth data is sample data with diversity; A second data processing unit is configured to create and generate a corresponding first model based on the fourth data, and to mark and process the first data according to the first model to generate corresponding fifth data; wherein the first model is a deep learning model for automatic point sampling; and the fifth data is point sampling location information data; The third data generating unit is used to construct a second model corresponding to the measurement according to the fifth data, and generate corresponding sixth data based on the second model; wherein the second model is a measurement element model; and the sixth data is measurement data corresponding to the product to be measured.

8. According to claim 7, an AI automatic point sampling optimization processing system for image measuring instruments based on adaptive learning is characterized in that: The first data generating unit further includes: A first generating module is used to generate and obtain seventh data corresponding to the product to be measured and processed; wherein the seventh data is activation control data for starting to collect product image data; A first processing module, used for marking and processing the first data according to the seventh data, and recording and maintaining the second data; And / or, the second data generating unit further includes: The first calculation module is used to calculate and generate third data corresponding to the second data based on the CSP backbone network structure; wherein the calculation formula is as follows: (1) In the formula, is the fused feature; are feature maps at different stages; and / or, the first data processing unit further includes: The second processing module is used to dynamically adjust the anchor box size and loss function weight of the model in combination with the adaptive learning algorithm; wherein the anchor box size is calculated as follows: (2) In the formula, For the Anchor box size; is the prediction box; is the real frame; is the intersection and comparison; A second generating module is used to process the second data in real time based on data enhancement technology, and generate corresponding eighth data; wherein the processing includes random scaling processing, cropping processing, flipping processing and translation processing; the eighth data is diversified sample data corresponding to the sample data; And / or, the second data processing unit further includes: The second calculation module is used to calculate and generate corresponding fifth data based on the first model and in combination with the first data; wherein the calculation formula is as follows: (3) In the formula, For automatic sampling point location; For location Confidence level; For location With real frame The intersection and union ratio of is the area to be sampled in the image; And / or, the third data generating unit further includes: A first construction module, configured to construct at least one corresponding measurement element according to the fifth data; wherein the measurement element includes a line, an arc and a distance; The third processing module is used to combine the stochastic gradient descent or Adam optimizer to respectively train and optimize the target detection model or the key point detection model based on the convolutional neural network.

9. An AI automatic point sampling optimization processing platform for image measuring instruments based on adaptive learning, characterized in that: It includes a processor, a memory, and an image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning; wherein, the image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning is executed in the processor, and the image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning is stored in the memory, and the image measuring instrument AI automatic point sampling optimization processing platform control program based on adaptive learning implements the image measuring instrument AI automatic point sampling optimization processing method based on adaptive learning as described in any one of claims 1 to 6.

Images