Edge Gateway System for AI Visual Recognition Processing Based on Large Model Scenario Application

Abstract

The present invention discloses an edge gateway system for AI visual recognition processing based on large model scenario applications, including a data acquisition module, a communication conversion module, an identification processing module, and a model application optimization module. The present invention relates to the field of artificial intelligence technology. The edge gateway system for AI visual recognition processing based on large model scenario applications preprocesses data, then restores the original features through visual recognition, conducts local feature analysis and extraction based on the original features, obtains the matching of the results after the recognition operation with the data in the database, and finally determines the target features, thereby using AI visual recognition processing to restore the analysis and restoration of abnormal and oblique features, and realizing the matching processing of the restored features, so as to effectively improve the accuracy and recognition efficiency in data processing, enabling devices with the edge gateway system to better realize the application of large model scenarios.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and specifically to an edge gateway system for AI visual recognition processing based on large model scenario applications. Background Art

[0002] With the development of artificial intelligence technology, large models have been widely applied in many fields. At the same time, AI visual recognition technology has become increasingly mature and plays an important role in multiple scenarios such as security, intelligent transportation, and industrial inspection.

[0003] Refer to the patent name: Edge Gateway System, Data Processing Method of Edge Gateway System (Patent Publication No.: CN118972433A, Patent Publication Date: November 15, 2024). The system includes at least one edge gateway device. The edge gateway device includes a multi-protocol access unit, an edge computing processing unit, and an external interface unit. The external interface unit is used to connect to terminal devices, and the multi-protocol access unit is used to connect to cloud devices. The edge computing processing unit includes a central processor and an auxiliary computing processor for performing auxiliary calculations on raw service data. The auxiliary computing processor is connected to the central processor. The central processor is connected to the external interface unit to communicate with terminal devices through the external interface unit. The central processor is connected to the multi-protocol access unit to communicate with cloud devices through the multi-protocol access unit. This system helps to reduce the cloud load, and helps to improve the data processing efficiency of the edge gateway system and the service processing ability of the edge gateway device.

[0004] Based on the description of the above document, in the existing edge gateway system at the toll road exit, during the recognition and monitoring of the license plate of a driving vehicle, the recognized characters have a deviation in direction, which easily leads to the problem that the recognized and processed data does not match the actual situation, or the processed data is still inaccurate, so that the final judgment result has an error, and even the problem of not recognizing the required data occurs. Therefore, the present invention provides an edge gateway system for AI visual recognition processing based on large model scenario applications. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides an edge gateway system for AI visual recognition processing based on large model scenario applications, which solves the problem that in the existing edge gateway system at the toll road exit, during the recognition and monitoring of the license plate of a driving vehicle, the recognized characters have a deviation in direction, which easily leads to the problem that the recognized and processed data does not match the actual situation, or the processed data is still inaccurate, so that the final judgment result has an error, and even the problem of not recognizing the required data occurs.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: An edge gateway system for AI visual recognition processing based on large model scenario applications, including:

[0007] A data acquisition module, equipped with sensors and cameras to implement image acquisition operations on physical objects in the scenario;

[0008] A communication conversion module, which performs format conversion and encoding processing on the collected data and establishes a transmission channel with the edge gateway device;

[0009] An identification processing module, which preprocesses the data, then restores the original features through visual recognition, performs local feature analysis and extraction based on the original features, and matches the results after the identification operation with the data in the database to finally determine the target features;

[0010] A model application optimization module, which summarizes and introduces the data of each identification process into a set model for training and optimization to realize the actual application of the trained model.

[0011] Preferably, the operation steps of the communication conversion module are as follows:

[0012] A1. Identify the communication protocol type of the collected data;

[0013] A2. According to the communication protocol requirements of the edge gateway device, perform format conversion and encoding processing operations on the collected data;

[0014] A3. Verify whether the converted data meets the protocol standards of the target system, and continue the conversion process until the protocol standards are the same, and then establish a transmission channel with the edge gateway device.

[0015] Preferably, the preprocessing operation of the identification processing module on the data is as follows:

[0016] B1. After receiving the collected data, perform image data extraction operations according to the set frame number period, and then screen the image data;

[0017] B2. Locate the license plate by extracting the features in the image, determine the license plate features according to the proportional value of the square license plate in the historical data. If no license plate features are determined in all images, generate a direction adjustment instruction and transmit it to the adjustment unit to realize the adjustment operation of the camera, and complete the preservation operation of the license plate features;

[0018] B3. And summarize the retained license plate feature image data and the parameter data at the time of current extraction and collection to form an initial data set labeled as P.

[0019] Preferably, the operation of the identification processing module to restore the original features through visual recognition is as follows:

[0020] C1. Extract the image data with license plate features and perform an amplification operation on the license plate features in the image data;

[0021] C2. Based on the edge boundaries of the license plate features, measure the vertical and horizontal distance change values. After marking the character features in the features, perform a reduction operation on the character features according to the change values to form image data with positive license plate features;

[0022] C3. Perform local analysis and extraction operations on the character features to confirm the actual situation of the features.

[0023] Preferably, the operation of amplifying the license plate features in C1 is as follows:

[0024] c11. Determine the horizontal boundary and vertical boundary of the license plate features in the image data, and the sizes of the horizontal boundary and vertical boundary are L1 and L2 respectively;

[0025] c12. Determine the sizes of the horizontal boundary and vertical boundary of the image data and label them as H1 and H2, and calculate the amplification ratio through the ratio of the corresponding horizontal boundary and vertical boundary;

[0026] c13. Compare the ratio of H1 / L1 with the ratio of H2 / L2, select the smaller ratio as the amplification ratio, and transmit the amplification ratio to the adjustment unit to implement the adjustment operation of the camera.

[0027] Preferably, the adjustment unit includes a direction adjustment group and a focal length adjustment group. The direction adjustment group implements the adjustment operation in the left-right or up-down direction after receiving the adjustment instruction until an image with license plate features appears. The focal length adjustment group implements the adjustment of the focal length after receiving the adjustment instruction to complete the amplification operation of the license plate features.

[0028] Preferably, the operation of reducing the character features in C2 is as follows:

[0029] c21. Determine the actual horizontal distance of the license plate features as L3, the actual vertical distance of the license plate features as L4, the horizontal distance of the license plate features in the image data as H3, and the vertical distance of the license plate features in the image data as H4, and determine the angular deviation in the horizontal direction labeled as α and the angular deviation in the vertical direction labeled as β;

[0030] c22. Then perform grayscale processing on the license plate features, and then distinguish the character features according to the grayscale values to mark the character features;

[0031] c23. Perform a reduction operation on each character feature, determine the size of the character feature in the image data, and calculate the actual character feature reduction value according to the angular deviation to form image data with positive license plate features.

[0032] Preferably, the angle deviation calculation operation in c21 is as follows:

[0033] Horizontal angle deviation: Sinα = H3 / L3;

[0034] Vertical angle deviation: Cosβ = H4 / L4.

[0035] Preferably, the operation of calculating the actual character feature restoration value based on the angle deviation in c23 is as follows:

[0036] D1. Select a character feature, and determine the upper, lower, left, and right extreme points of the current character feature. Connect the upper and lower extreme points and the left and right extreme points, and the intersection point of the two connection lines is the center point of the character feature;

[0037] D2. Establish an X / Y-axis cross coordinate system with the center point as the origin, and equidistantly set multiple line segments from the upper to the lower extreme points. Mark the intersection points of the line segments and the boundary of the character feature as expansion nodes. Expand based on the distance from the expansion nodes to the Y-axis according to the horizontal angle deviation value, and restore and connect the expanded nodes according to the situation of the character feature;

[0038] Set the distance M1 from the expansion node to the Y-axis, and the corresponding actual horizontal distance is N1. The calculation formula for the actual horizontal distance N1 is: N1 = M1 / (H3 / L3);

[0039] D3. Then, based on the character feature after connection in D2, equidistantly set multiple line segments from the left to the right extreme points. Mark the intersection points of the line segments and the boundary of the character feature as expansion nodes. Expand based on the distance from the expansion nodes to the X-axis according to the vertical angle deviation value, and restore and connect the expanded nodes according to the situation of the character feature to obtain the character feature of the original size after restoration;

[0040] Set the distance M2 from the expansion node to the X-axis, and the corresponding actual horizontal distance is N2. The calculation formula for the actual horizontal distance N2 is: N2 = M2 / (H4 / L4).

[0041] Preferably, the operation steps for local analysis and extraction of character features in C3 are as follows:

[0042] c31. After extracting the restored character feature, extract each local feature on the character feature;

[0043] c32. Extract the local features corresponding to similar characters from the historical database, and match and compare the local features of the historical database with the local features of the character feature;

[0044] c33. Determine the result of the corresponding abnormal character feature according to the matching situation.

[0045] The present invention provides an edge gateway system for AI visual recognition processing based on large model scenario applications. Compared with the prior art, it has the following beneficial effects:

[0046] 1. For the edge gateway system for AI visual recognition processing based on large model scenario applications, after preprocessing the data, the original features are restored through visual recognition, local feature analysis and extraction are performed based on the original features, the result after the recognition operation is matched with the data in the database, and finally the target features are determined. Thus, the analysis and restoration of abnormal and oblique features are restored by using AI visual recognition processing, and the matching process of the restored features is realized, effectively improving the accuracy and recognition efficiency in data processing, and enabling the devices with the edge gateway system to better realize the application of large model scenarios.

[0047] 2. For the edge gateway system for AI visual recognition processing based on large model scenario applications, by extracting the image data with license plate features, the horizontal and vertical boundaries of the license plate features in the image data and the actual ones are determined, the magnification ratio is calculated by the ratio of the corresponding horizontal and vertical boundaries, and the smaller ratio is selected to perform the magnification processing operation on the license plate features in the image data. Thus, while maintaining the pixels, the magnification processing of the collection of license plate features is carried out, the accuracy of data collection is improved, and it is associated with the subsequent processing algorithm to ensure the efficiency of data processing.

[0048] 3. For the edge gateway system for AI visual recognition processing based on large model scenario applications, by performing the restoration operation on each character feature, the size of the character feature located in the image data is determined, and the actual character feature restoration value is calculated based on the angle deviation to form the image data facing the license plate features. Thus, adaptive adjustment is made according to the situations of different vehicles to realize the restoration of the character features facing the image data, which is convenient for better subsequent matching and comparison of local features and ensures the accuracy of the recognition processing operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is the principle block diagram of the edge gateway system of the present invention;

[0050] Figure 2 is the operation flowchart of the communication conversion module of the present invention;

[0051] Figure 3 is the preprocessing operation flowchart of the recognition processing module of the present invention;

[0052] Figure 4 is the visual recognition restoration feature operation flowchart of the recognition processing module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0054] Please refer to Figures 1 - 4 , the present invention provides two technical solutions:

[0055] Embodiment 1. An edge gateway system for AI visual recognition processing based on large model scenario applications, including:

[0056] A data acquisition module equipped with sensors and cameras to perform image acquisition operations on physical objects in the scenario;

[0057] A communication conversion module that performs format conversion and encoding processing on the acquired data and establishes a transmission channel with the edge gateway device;

[0058] An identification processing module that preprocesses the data, then restores the original features through visual recognition, performs local feature analysis and extraction based on the original features, obtains the matching of the results after the identification operation with the data in the database, and finally determines the target features;

[0059] A model application optimization module that summarizes and introduces the data of each identification process into a set model for training and optimization to realize the actual application of the trained model.

[0060] Among them, by preprocessing the data, then restoring the original features through visual recognition, performing local feature analysis and extraction based on the original features, obtaining the matching of the results after the identification operation with the data in the database, and finally determining the target features, thereby using AI visual recognition processing to restore the analysis and restoration of abnormal and oblique features, and realizing the matching processing of the restored features, so as to effectively improve the accuracy and recognition efficiency in data processing, and enable the device with the edge gateway system to better realize the application of the large model scenario.

[0061] The edge gateway for physical object acquisition in the scenario includes a camera and an audible and visual alarm, so that when the system fails to complete the identification process operation, the audible and visual alarm can inform the personnel to handle it, reducing the time affected after the failure occurs.

[0062] In the embodiment of the present invention, the operation steps of the communication conversion module are:

[0063] A1. Identify the communication protocol type of the acquired data;

[0064] A2. According to the communication protocol requirements of the edge gateway device, perform operations on format conversion and encoding processing of the collected data;

[0065] A3. Verify whether the converted data meets the protocol standards of the target system, and continue the conversion process until the protocol standards are the same, and then establish a transmission channel with the edge gateway device.

[0066] In the embodiment of the present invention, the preprocessing operation of the recognition processing module on the data is as follows:

[0067] B1. After receiving the collected data, perform extraction operation on the picture data according to the set frame number period, and then screen the picture data;

[0068] B2. Perform license plate positioning by extracting the features in the image, determine the license plate features according to the proportional value of the square license plate in the historical data. If the license plate features are not determined in all images, generate a direction adjustment instruction and transmit it to the adjustment unit to implement the adjustment operation of the camera, and complete the saving operation of the license plate features;

[0069] B3. Summarize the retained license plate feature image data and the parameter data at the time of current extraction and collection to form an initial data set labeled as P.

[0070] In the embodiment of the present invention, the operation of the recognition processing module to restore the original features through visual recognition is as follows:

[0071] C1. Extract the image data with license plate features, and perform an amplification processing operation on the license plate features in the image data;

[0072] C2. Based on the edge boundaries of the license plate features, measure the vertical and horizontal distance change values. After marking the character features in the features, perform the restoration operation on the character features according to the change values to form image data with positive license plate features;

[0073] C3. Perform local analysis and extraction operations on the character features to confirm the actual situation of the features.

[0074] In the embodiment of the present invention, the operation of amplifying the license plate features in C1 is as follows:

[0075] c11. Determine the horizontal boundary and vertical boundary of the license plate features in the image data, and the sizes of the horizontal boundary and vertical boundary are L1 and L2 respectively;

[0076] c12. Determine the sizes of the horizontal boundary and vertical boundary of the image data and label them as H1 and H2, and calculate the amplification ratio through the ratio of the corresponding horizontal boundary and vertical boundary;

[0077] c13. Compare the ratio of H1 / L1 with the ratio of H2 / L2, select the smaller ratio as the magnification ratio, and transmit the magnification ratio to the adjustment unit to implement the adjustment operation of the camera.

[0078] Among them, by extracting the image data with license plate features, determining the horizontal and vertical boundaries of the license plate features in the image data and the actual license plate features, calculating and determining the magnification ratio through the ratio of the corresponding horizontal and vertical boundaries, and selecting the smaller ratio to perform the magnification processing operation on the license plate features in the image data, so as to perform the magnification processing on the acquisition of the license plate features while maintaining the pixels, improve the accuracy of data acquisition, and be associated with the subsequent processing algorithm to ensure the efficiency of data processing.

[0079] In the embodiment of the present invention, the adjustment unit includes a direction adjustment group and a focal length adjustment group. The direction adjustment group implements the adjustment operation in the left-right or up-down direction after receiving the adjustment instruction until an image with license plate features appears. The focal length adjustment group implements the adjustment of the focal length after receiving the adjustment instruction to complete the magnification operation of the license plate features.

[0080] Among them, the direction adjustment group and the focal length adjustment group are mature and existing devices used to implement the adjustment control operation after being combined with the edge gateway system. Any device capable of completing the direction adjustment and focal length adjustment can be used, and no detailed description will be given in this article.

[0081] In the embodiment of the present invention, the operation of restoring the character features in C2 is as follows:

[0082] c21. Determine that the actual horizontal distance of the license plate feature is L3, the actual vertical distance of the license plate feature is L4, the horizontal distance of the license plate feature in the image data is H3, and the vertical distance of the license plate feature in the image data is H4, and determine the angle deviation α in the horizontal direction and the angle deviation β in the vertical direction;

[0083] c22. Then perform grayscale processing on the license plate features, and then distinguish the character features according to the grayscale values to implement the marking of the character features;

[0084] c23. Perform the restoration operation on each character feature, determine the size of the character feature in the image data, and calculate the actual character feature restoration value according to the angle deviation to form the image data facing the license plate feature.

[0085] Among them, after grayscale processing, the grayscale values corresponding to different features also have differences. Set multiple detection points equidistantly on the image data, determine the position of the character features according to the change of the grayscale values of the detection points, and then extract the first numerical change detection points located outside the character features and connect them in sequence to form the outer boundary of the character features.

[0086] In the embodiment of the present invention, the operation of calculating the angular deviation in c21 is as follows:

[0087] Horizontal angular deviation: Sinα = H3 / L3;

[0088] Vertical angular deviation: Cosβ = H4 / L4.

[0089] In the embodiment of the present invention, the operation of calculating the actual character feature restoration value based on the angular deviation in c23 is as follows:

[0090] D1. Select a character feature, and determine the upper, lower, left, and right extreme points of the current character feature. Connect the upper and lower extreme points and the left and right extreme points, and the intersection point of the two connecting lines is the center point of the character feature;

[0091] D2. Establish an X / Y-axis cross coordinate system with the center point as the origin, and equidistantly set multiple line segments from the upper to the lower extreme points. Mark the intersection points of the line segments and the boundary of the character feature as expansion nodes. Expand based on the distance from the expansion node to the Y-axis according to the horizontal angular deviation value, and restore and connect the expanded nodes according to the situation of the character feature;

[0092] Set the distance M1 from the expansion node to the Y-axis, and the corresponding actual horizontal distance is N1. The calculation formula for the actual horizontal distance N1 is: N1 = M1 / (H3 / L3);

[0093] D3. Then, based on the character feature after connection in D2, equidistantly set multiple line segments from the left to the right extreme points. Mark the intersection points of the line segments and the boundary of the character feature as expansion nodes. Expand based on the distance from the expansion node to the X-axis according to the vertical angular deviation value, and restore and connect the expanded nodes according to the situation of the character feature to obtain the restored character feature of the original size;

[0094] Set the distance M2 from the expansion node to the X-axis, and the corresponding actual horizontal distance is N2. The calculation formula for the actual horizontal distance N2 is: N2 = M2 / (H4 / L4).

[0095] Among them, by performing a restoration operation on each character feature, the size of the character feature located in the image data is determined, and the actual character feature restoration value is calculated based on the angular deviation to form image data facing the license plate feature, so as to make an adaptive adjustment according to the conditions of different vehicles, realize the restoration of the character feature facing the image data, and thus facilitate better subsequent matching and comparison of local features and ensure the accuracy of the recognition and processing operation.

[0096] In the embodiment of the present invention, the operation steps of locally analyzing and extracting the character feature in C3 are as follows:

[0097] c31. After extracting the restored character features, extract various local features on the character features;

[0098] c32. Extract the local features corresponding to similar characters from the historical database, and match and compare the local features of the historical database with the local features of the character features;

[0099] c33. Determine the result corresponding to the abnormal character features according to the matching situation.

[0100] The difference in the second embodiment compared with the first embodiment is that in the local analysis operation of character features, for example, when distinguishing and recognizing the character features "2" and "Z", through the recognition of local features, the upper half of the character feature "2" is arc-shaped, while the upper half of the character feature "Z" is a combination of a straight line and a corner. Therefore, after matching and comparing with the local features of the corresponding restored character features as distinguishing features, the actual character features are confirmed;

[0101] When distinguishing and recognizing the character features "D" and "O", through the recognition of local features, the left half of the character feature "D" is vertical, while the left half of the character feature "O" is arc-shaped. Therefore, after matching and comparing with the local features of the corresponding restored character features as distinguishing features, the actual character features are confirmed.

[0102] At the same time, the content not described in detail in this specification belongs to the prior art well known to those skilled in the art.

[0103] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.

[0104] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An edge gateway system for AI visual recognition processing based on large model scene applications, characterized by: include: The data acquisition module is equipped with sensors and cameras to realize image acquisition operations of physical objects in the scene; The communication conversion module performs format conversion and encoding processing on the collected data and establishes a transmission channel with the edge gateway device; The recognition processing module pre-processes the data, then restores the original features through visual recognition, performs local feature analysis and extraction based on the original features, matches the results of the recognition operation with the data in the database, and finally determines the target features; The model application optimization module summarizes the target feature data of each recognition process and introduces it into the set model for training and optimization to realize the practical application of the trained model; The operation of restoring the original features by visual recognition in the recognition processing module is as follows: C1. Extracting image data with license plate features, and performing a magnification operation on the license plate features in the image data; C2, and based on the edge limit of the license plate feature, the vertical and horizontal distance change values ​​are calculated, and after marking the character features in the feature, the character features are restored according to the change value to form image data with the license plate feature; C3, perform local analysis and extraction operations on character features to confirm the actual situation of the features; The operation of restoring character features in C2 is: c21. Determine that the actual lateral distance of the license plate feature is L3, the actual vertical distance of the license plate feature is L4, the lateral distance of the license plate feature in the image data is H3, the vertical distance of the license plate feature in the image data is H4, and determine the lateral angle deviation α and the vertical angle deviation β; c22, then grayscale the license plate features, and then distinguish the character features according to the grayscale value to achieve the marking of the character features; c23, performing a restoration operation on each character feature, determining the size of the character feature in the image data, and calculating the actual character feature restoration value based on the angle deviation to form image data directly facing the license plate feature; The operation of calculating the actual character feature restoration value based on the angle deviation in c23 is: Determine the upper, lower, left, and right extreme points of the character feature, connect the upper and lower extreme points and the left and right extreme points, and the intersection of the two lines is the center point of the character feature; An X / Y axis cross coordinate system is established with the center point as the origin, and the expansion node is determined based on the extreme point. The expansion is performed according to the angle deviation value based on the distance from the expansion node to the coordinate axis, and the expanded nodes are restored and connected according to the character features.

2. The edge gateway system for AI visual recognition processing based on large model scene application according to claim 1 is characterized in that: The operation steps of the communication conversion module are: A1. Identify the communication protocol type of the collected data; A2. Perform format conversion and encoding processing operations on the collected data according to the communication protocol requirements of the edge gateway device; A3. Verify whether the converted data meets the protocol standard of the target system, continue the conversion process until the protocol standard is the same, and then establish a transmission channel with the edge gateway device.

3. The edge gateway system for AI visual recognition processing based on large model scene application according to claim 1 is characterized in that: The identification processing module performs preprocessing operations on the data as follows: B1. After receiving the collected data, extract the image data according to the set frame period, and then screen the image data; B2. The license plate is located by extracting features from the image, and the license plate features are determined by the ratio of the square license plate in the historical data. If the license plate features are not determined in all images, a direction adjustment instruction is generated and transmitted to the adjustment unit to implement the adjustment operation of the camera, and the image data with the license plate features is saved; B3, and summarize the retained license plate feature image data and the parameter data currently extracted and collected to form an initial data set marked as P.

4. The edge gateway system for AI visual recognition processing based on large model scene application according to claim 1 is characterized in that: The operation of enlarging the license plate feature in C1 is: c11, determining the horizontal boundary and the vertical boundary of the license plate feature in the image data, and the sizes of the horizontal boundary and the vertical boundary are L1 and L2 respectively; c12, determining the sizes of the horizontal boundary and the vertical boundary of the image data and marking them as H1 and H2, and calculating the magnification ratio by the ratio of the corresponding horizontal boundary and the vertical boundary; c13. Compare the ratio of H1 / L1 with the ratio of H2 / L2, select the smaller ratio as the magnification ratio, and transmit the magnification ratio to the adjustment unit to implement the adjustment operation of the camera.

5. The edge gateway system for AI visual recognition processing based on large model scene application according to claim 4 is characterized in that: The adjustment group includes a direction adjustment group and a focal length adjustment group, and the direction adjustment group implements left-right or up-down adjustment operations after receiving an adjustment instruction until an image with license plate features appears, and the focal length adjustment group implements focal length adjustment to complete the magnification operation of the license plate features after receiving an adjustment instruction.

6. The edge gateway system for AI visual recognition processing based on large model scene application according to claim 1 is characterized in that: The angle deviation calculation operation in c21 is: Lateral angular deviation: Sinα=H3 / L3; Vertical angular deviation: Cosβ=H4 / L4.

7. The edge gateway system for AI visual recognition processing based on large model scene application according to claim 6 is characterized in that: The operation of calculating the actual character feature restoration value based on the angle deviation in c23 is: D1. Select a character feature and determine the upper, lower, left and right extreme points of the current character feature. Connect the upper and lower extreme points and the left and right extreme points, and the intersection of the two lines is the center point of the character feature. D2. Establish an X / Y axis cross coordinate system with the center point as the origin, and set multiple lines at equal distances from the top to the bottom extreme point, and mark the intersection of the line and the boundary of the character feature as an expansion node. Based on the distance from the expansion node to the Y axis, expand according to the horizontal angle deviation value, and restore the expanded nodes according to the character feature. The distance from the expansion node to the Y axis is set to M1, and the corresponding actual horizontal distance is N1, and the calculation formula of the actual horizontal distance N1 is: N1=M1 / (H3 / L3); D3, then on the character features connected in D2, multiple lines are set at equal distances from the left to the right extreme points, and the intersection of the lines and the boundaries of the character features is marked as an expansion node, and the expansion is performed according to the vertical angle deviation value based on the distance from the expansion node to the X-axis, and the expanded nodes are restored and connected according to the situation of the character features to obtain the character features of the original size after restoration; The distance from the expansion node to the X-axis is set to M2, and the corresponding actual lateral distance is N2, and the calculation formula of the actual lateral distance N2 is: N2=M2 / (H4 / L4).

8. The edge gateway system for AI visual recognition processing based on large model scene application according to claim 1 is characterized in that: The steps of performing local analysis and extraction of character features in C3 are as follows: c31. After extracting the restored character features, extract various local features on the character features; c32. extracting local features corresponding to similar characters from the historical database, and matching and comparing the local features of the historical database with the local features of the character features; c33. Determine the result corresponding to the abnormal character feature according to the matching situation.

Images