License plate recognition image data analysis processing method based on digital watermarking

Abstract

The invention relates to the field of license plate recognition image data analysis and processing, in particular to a license plate recognition image data analysis and processing method based on digital watermarking, which comprises the following steps of: performing feature region division processing by using a real-time license plate recognition image to respectively obtain a main feature region and a secondary feature region of the real-time license plate recognition image; performing target content recognition processing by using the main feature region of the real-time license plate recognition image to obtain a target content recognition processing result of the main feature region; using the target content identification processing result, according to the secondary feature area of the real-time license plate identification image, carrying out cross validation processing based on digital watermarking, and obtaining a cross validation processing result of the real-time license plate identification image; obtaining a license plate recognition image data analysis processing result by using a cross validation processing result of the real-time license plate recognition image; the license plate image is divided into a plurality of areas in detail, so that the final processing result of the license plate image data can be quickly and effectively obtained.

Description

Technical Field

[0001] The invention relates to the field of license plate recognition image data analysis and processing, and in particular to a license plate recognition image data analysis and processing method based on digital watermark. Background Art

[0002] As its core component, license plate recognition technology has been widely used in traffic management, vehicle monitoring, parking lot management and other fields. License plate recognition technology mainly extracts license plate information from vehicle images through image processing, pattern recognition and machine learning, and then identifies and verifies it. However, existing license plate recognition technology still faces many challenges in practical applications, mainly including the following aspects:

[0003] Insufficient recognition accuracy in complex environments:

[0004] In actual applications, license plate recognition systems are often affected by factors such as lighting changes, weather conditions, license plate damage, and occlusion, which results in a decrease in recognition accuracy. For example, at night or in rainy and snowy weather, the image quality is poor, and traditional image processing algorithms have difficulty accurately extracting license plate area and character information;

[0005] License plate forgery and tampering issues:

[0006] With the popularization of license plate recognition technology, some criminals evade traffic supervision by forging and tampering with license plate information. Most existing license plate recognition systems rely only on the recognition of license plate characters and lack an effective verification mechanism for the authenticity of license plates, making it difficult to deal with forgery and tampering.

[0007] Insufficient reliability of recognition results:

[0008] Traditional license plate recognition systems usually use a single recognition algorithm and lack a cross-validation mechanism for recognition results. When errors occur in the recognition algorithm, the system cannot automatically correct them, resulting in the direct output of erroneous results, which affects the reliability and practicality of the system.

[0009] Inefficient data processing:

[0010] In real-time license plate recognition scenarios, the system needs to quickly process a large amount of image data. However, existing license plate recognition technologies often require a long computing time when processing complex images, making it difficult to meet real-time requirements;

[0011] Therefore, there is an urgent need for an efficient and reliable license plate recognition method that can achieve high-precision license plate recognition in complex environments and ensure the accuracy and authenticity of the recognition results through an effective verification mechanism. Summary of the invention

[0012] In view of the shortcomings of the prior art, the present invention provides a method for analyzing and processing license plate recognition image data based on digital watermarks, which realizes rapid and accurate analysis and processing of image data by regional classification processing of real-time license plate recognition images.

[0013] To achieve the above object, the present invention provides a method for analyzing and processing license plate recognition image data based on digital watermark, comprising:

[0014] S1. Perform feature region division processing using the real-time license plate recognition image to obtain the main feature region and the secondary feature region of the real-time license plate recognition image;

[0015] S2, performing target content recognition processing using the main feature area of ​​the real-time license plate recognition image to obtain a target content recognition processing result of the main feature area;

[0016] S3, using the target content recognition processing result, according to the secondary feature area of ​​the real-time license plate recognition image, cross-validation processing is performed based on the digital watermark to obtain a cross-validation processing result of the real-time license plate recognition image;

[0017] S4. Using the cross-validation processing result of the real-time license plate recognition image, obtain the license plate recognition image data analysis processing result.

[0018] Preferably, the feature area division process using the real-time license plate recognition image to obtain the main feature area and the secondary feature area of ​​the real-time license plate recognition image includes:

[0019] Using the real-time license plate recognition image to establish a pixel matrix of the real-time license plate recognition image;

[0020] Performing grayscale value conversion processing using the real-time license plate recognition image to obtain a real-time license plate recognition grayscale image;

[0021] Use image segmentation algorithms or deep learning models to process real-time license plate recognition grayscale images and extract primary and secondary feature areas;

[0022] The main feature area includes the license plate number area, and the secondary feature area includes the license plate frame, color area, background area or special logo area;

[0023] The extracted main feature area and secondary feature area are optimized to obtain the main feature area and secondary feature area of ​​the real-time license plate recognition image.

[0024] Furthermore, the target content recognition processing is performed using the main feature area of ​​the real-time license plate recognition image to obtain the target content recognition processing result of the main feature area, including:

[0025] Use optical character recognition technology to identify the main feature area, extract the license plate number information to obtain the license plate information recognition result;

[0026] Correcting and optimizing the license plate information recognition result to obtain a corrected and optimized result of the license plate information recognition result;

[0027] Utilizing the correction optimization result of the license plate information recognition result to obtain the target content recognition processing result of the main feature area based on the convolutional neural network model;

[0028] The correction and optimization processing includes character segmentation, character matching and error correction.

[0029] Furthermore, the cross-validation processing is performed based on the digital watermark according to the secondary feature area of ​​the real-time license plate recognition image using the target content recognition processing result to obtain the cross-validation processing result of the real-time license plate recognition image, including:

[0030] Embedding digital watermark information in the secondary feature area of ​​the real-time license plate recognition image based on discrete cosine transform;

[0031] Acquire digital watermark information and compare it with the target content recognition processing result to obtain a real-time comparison processing result;

[0032] Determine whether the digital watermark information corresponding to the real-time comparison processing result is consistent with the target content recognition processing result. If so, the cross-validation processing result of the real-time license plate recognition image is passed; otherwise, the cross-validation processing result of the real-time license plate recognition image is failed;

[0033] The digital watermark information includes a license plate number, a timestamp, an encrypted hash value and a unique identifier.

[0034] Furthermore, the cross-validation processing result of the real-time license plate recognition image is used to obtain the license plate recognition image data analysis processing result, including:

[0035] Obtaining a final recognition result of the license plate number using the cross-validation processing result of the real-time license plate recognition image;

[0036] Utilizing the cross-validation processing result of the real-time license plate recognition image, and calculating the confidence score of the license plate recognition based on the target content recognition processing result and the digital watermark verification result;

[0037] Obtaining a verification status of the license plate recognition using a cross-validation processing result of the real-time license plate recognition image;

[0038] The timestamp, geographic location information and environmental parameters of the license plate recognition are obtained using the cross-validation processing result of the real-time license plate recognition image.

[0039] Furthermore, it also includes:

[0040] For license plate numbers that fail verification, feature information is re-extracted based on the secondary feature area and compared with the target content recognition processing result;

[0041] Backtrack the problematic steps (such as image segmentation, character recognition or digital watermark extraction) and re-execute the relevant processing;

[0042] Self-inspection and self-correction are achieved through an internal loop iteration mechanism to ensure the accuracy and stability of the final license plate image recognition results.

[0043] Furthermore, the internal loop iteration mechanism includes:

[0044] In each iteration, the processing results and intermediate data of the current step are recorded;

[0045] Analyze the root cause of the problem based on the recorded data and adjust relevant parameters or algorithms;

[0046] Repeat steps S1 to S4 until the license plate recognition result is verified or the preset number of iterations is reached;

[0047] If the verification fails after reaching the preset number of iterations, an exception report will be output and manual intervention will be prompted.

[0048] Further, the outputting of the abnormality report and prompting for manual intervention includes:

[0049] During the target content identification and cross-validation process, the processing results and intermediate data of each step are recorded for subsequent analysis and optimization;

[0050] License plate numbers that fail verification are marked and an exception report is generated to prompt manual intervention or further processing.

[0051] Compared with the closest prior art, the present invention has the following beneficial effects:

[0052] For the license plate images collected in real time, a regional feature processing method is adopted. By carefully dividing the license plate images into multiple regions and accurately extracting the features in each region, the final processing results of the license plate image data can be obtained quickly and effectively. This process not only improves the speed of data processing, but also ensures the accuracy of the results.

[0053] In addition, the solution specially designs an internal loop iteration mechanism, which can automatically adjust and optimize the processing flow according to changes in the environment and conditions after actual deployment. This adaptive feature enables the system to maintain high applicability and efficiency in different application scenarios. Through this intelligent processing method, the license plate recognition system can not only meet current recognition needs, but also has good scalability and future adaptability, providing strong technical support for intelligent traffic management. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 The present invention provides a flowchart of a method for analyzing and processing license plate recognition image data based on digital watermark. DETAILED DESCRIPTION

[0055] The specific implementation modes of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0056] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Embodiment 1:

[0058] The present invention provides a method for analyzing and processing license plate recognition image data based on digital watermark, such as Figure 1 As shown, including:

[0059] S1. Perform feature region division processing using the real-time license plate recognition image to obtain the main feature region and the secondary feature region of the real-time license plate recognition image;

[0060] S2, performing target content recognition processing using the main feature area of ​​the real-time license plate recognition image to obtain a target content recognition processing result of the main feature area;

[0061] S3, using the target content recognition processing result, according to the secondary feature area of ​​the real-time license plate recognition image, cross-validation processing is performed based on the digital watermark to obtain a cross-validation processing result of the real-time license plate recognition image;

[0062] S4. Using the cross-validation processing result of the real-time license plate recognition image, obtain the license plate recognition image data analysis processing result.

[0063] S1 specifically includes:

[0064] S1-1, using the real-time license plate recognition image to establish a pixel matrix of the real-time license plate recognition image;

[0065] S1-2, using the real-time license plate recognition image to perform grayscale value conversion processing to obtain a real-time license plate recognition grayscale image;

[0066] S1-3, use image segmentation algorithm or deep learning model to process the real-time license plate recognition grayscale image and extract the main feature area and secondary feature area;

[0067] S1-4, the main feature area includes the license plate number area, and the secondary feature area includes the license plate frame, color area, background area or special logo area;

[0068] S1-5. Optimize the extracted main feature area and secondary feature area to obtain the main feature area and secondary feature area of ​​the real-time license plate recognition image.

[0069] S2 specifically includes:

[0070] S2-1, using optical character recognition technology to identify the main feature area, extract the license plate number information to obtain the license plate information recognition result;

[0071] S2-2, correcting and optimizing the license plate information recognition result to obtain a corrected and optimized result of the license plate information recognition result;

[0072] S2-3, using the correction optimization result of the license plate information recognition result to obtain the target content recognition processing result of the main feature area based on the convolutional neural network model;

[0073] The correction and optimization processing includes character segmentation, character matching and error correction.

[0074] S3 specifically includes:

[0075] S3-1, using the secondary feature area of ​​the real-time license plate recognition image to embed digital watermark information in the secondary feature area based on discrete cosine transform;

[0076] S3-2, obtaining the digital watermark information and comparing it with the target content recognition processing result to obtain a real-time comparison processing result;

[0077] S3-3, judging whether the digital watermark information corresponding to the real-time comparison processing result is consistent with the target content recognition processing result, if so, the cross-validation processing result of the real-time license plate recognition image is passed, otherwise, the cross-validation processing result of the real-time license plate recognition image is failed;

[0078] The digital watermark information includes a license plate number, a timestamp, an encrypted hash value and a unique identifier.

[0079] S4 specifically includes:

[0080] S4-1, obtaining a final recognition result of the license plate number using the cross-validation processing result of the real-time license plate recognition image;

[0081] S4-2, using the cross-validation processing result of the real-time license plate recognition image to calculate the confidence score of the license plate recognition based on the target content recognition processing result and the digital watermark verification result;

[0082] S4-3, obtaining the verification status of the license plate recognition using the cross-validation processing result of the real-time license plate recognition image;

[0083] S4-4. Obtain the timestamp, geographic location information and environmental parameters of the license plate recognition using the cross-validation processing result of the real-time license plate recognition image.

[0084] In this embodiment, a method for analyzing and processing license plate recognition image data based on digital watermarking further includes:

[0085] For license plate numbers that fail verification, feature information is re-extracted based on the secondary feature area and compared with the target content recognition processing result;

[0086] Backtrack the problematic steps (such as image segmentation, character recognition or digital watermark extraction) and re-execute the relevant processing;

[0087] Self-inspection and self-correction are achieved through an internal loop iteration mechanism to ensure the accuracy and stability of the final license plate image recognition results.

[0088] The internal loop iteration mechanism includes:

[0089] In each iteration, the processing results and intermediate data of the current step are recorded;

[0090] Analyze the root cause of the problem based on the recorded data and adjust relevant parameters or algorithms;

[0091] Repeat steps S1 to S4 until the license plate recognition result is verified or the preset number of iterations is reached;

[0092] If the verification fails after reaching the preset number of iterations, an exception report will be output and manual intervention will be prompted.

[0093] The outputting of abnormal report and prompting manual intervention includes:

[0094] During the target content identification and cross-validation process, the processing results and intermediate data of each step are recorded for subsequent analysis and optimization;

[0095] License plate numbers that fail verification are marked and an exception report is generated to prompt manual intervention or further processing.

[0096] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0097] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0098] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0099] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for analyzing and processing license plate recognition image data based on digital watermark, characterized in that: include: S1. Perform feature region division processing using the real-time license plate recognition image to obtain the main feature region and the secondary feature region of the real-time license plate recognition image; S2, performing target content recognition processing using the main feature area of ​​the real-time license plate recognition image to obtain a target content recognition processing result of the main feature area; S3, using the target content recognition processing result, according to the secondary feature area of ​​the real-time license plate recognition image, cross-validation processing is performed based on the digital watermark to obtain a cross-validation processing result of the real-time license plate recognition image; S4. Using the cross-validation processing result of the real-time license plate recognition image, obtain the license plate recognition image data analysis processing result.

2. The data analysis method based on real-time license plate recognition image according to claim 1 is characterized in that: The feature area division process using the real-time license plate recognition image to obtain the main feature area and the secondary feature area of ​​the real-time license plate recognition image includes: Using the real-time license plate recognition image to establish a pixel matrix of the real-time license plate recognition image; Performing grayscale value conversion processing using the real-time license plate recognition image to obtain a real-time license plate recognition grayscale image; Use image segmentation algorithms or deep learning models to process real-time license plate recognition grayscale images and extract primary and secondary feature areas; The main feature area includes the license plate number area, and the secondary feature area includes the license plate frame, color area, background area or special logo area; The extracted main feature area and secondary feature area are optimized to obtain the main feature area and secondary feature area of ​​the real-time license plate recognition image.

3. The data analysis method based on real-time license plate recognition image according to claim 2 is characterized in that: The target content recognition processing is performed using the main feature area of ​​the real-time license plate recognition image to obtain the target content recognition processing result of the main feature area, which includes: Use optical character recognition technology to identify the main feature area, extract the license plate number information to obtain the license plate information recognition result; Correcting and optimizing the license plate information recognition result to obtain a corrected and optimized result of the license plate information recognition result; Utilizing the correction optimization result of the license plate information recognition result to obtain the target content recognition processing result of the main feature area based on the convolutional neural network model; The correction and optimization processing includes character segmentation, character matching and error correction.

4. A data analysis method based on real-time license plate recognition images according to claim 3, characterized in that: The cross-validation processing is performed based on the digital watermark according to the secondary feature area of ​​the real-time license plate recognition image using the target content recognition processing result to obtain the cross-validation processing result of the real-time license plate recognition image, including: Embedding digital watermark information in the secondary feature area of ​​the real-time license plate recognition image based on discrete cosine transform; Acquire digital watermark information and compare it with the target content recognition processing result to obtain a real-time comparison processing result; Determine whether the digital watermark information corresponding to the real-time comparison processing result is consistent with the target content recognition processing result. If so, the cross-validation processing result of the real-time license plate recognition image is passed; otherwise, the cross-validation processing result of the real-time license plate recognition image is failed; The digital watermark information includes a license plate number, a timestamp, an encrypted hash value and a unique identifier.

5. The data analysis method based on real-time license plate recognition image according to claim 4 is characterized in that: The cross-validation processing result of the real-time license plate recognition image is used to obtain the license plate recognition image data analysis processing result, which includes: Obtaining a final recognition result of the license plate number using the cross-validation processing result of the real-time license plate recognition image; Utilizing the cross-validation processing result of the real-time license plate recognition image, and calculating the confidence score of the license plate recognition based on the target content recognition processing result and the digital watermark verification result; Obtaining a verification status of the license plate recognition using a cross-validation processing result of the real-time license plate recognition image; The timestamp, geographic location information and environmental parameters of the license plate recognition are obtained using the cross-validation processing result of the real-time license plate recognition image.

6. A data analysis method based on real-time license plate recognition images according to claim 5, characterized in that: Also includes: For license plate numbers that fail verification, feature information is re-extracted based on the secondary feature area and compared with the target content recognition processing result; Backtrack the problematic steps (such as image segmentation, character recognition or digital watermark extraction) and re-execute the relevant processing; Self-inspection and self-correction are achieved through an internal loop iteration mechanism to ensure the accuracy and stability of the final license plate image recognition results.

7. The data analysis method based on real-time license plate recognition image according to claim 6 is characterized in that: The internal loop iteration mechanism includes: In each iteration, the processing results and intermediate data of the current step are recorded; Analyze the root cause of the problem based on the recorded data and adjust relevant parameters or algorithms; Repeat steps S1 to S4 until the license plate recognition result is verified or the preset number of iterations is reached; If the verification fails after reaching the preset number of iterations, an exception report will be output and manual intervention will be prompted.

8. A data analysis method based on real-time license plate recognition images according to claim 7, characterized in that: The outputting of abnormal report and prompting manual intervention includes: During the target content identification and cross-validation process, the processing results and intermediate data of each step are recorded for subsequent analysis and optimization; License plate numbers that fail verification are marked and an exception report is generated to prompt manual intervention or further processing.

Images