A method and computer device for coding license plates and inkjet codes in car stickers
By combining differentiable binary algorithm and regression algorithm, automatic detection and coding of small and medium-sized license plates and ink coding of car stickers is realized, and the problem of insufficient detection capabilities of small license plates and ink coding in the existing technology is solved, and the fairness and accuracy of the detection are improved.
Patent Information
- Application Number
- CN202210182005.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-25
- Publication Date
- 2025-04-08
- Estimated Expiration
- 2042-02-25
AI Technical Summary
The prior art is difficult to effectively detect and code small license plates and ink coding, and it lacks the ability to detect ink coding.
Using a combination of differentiable binary algorithm and regression algorithm, the license plate and ink code in the car sticker are automatically detected and coded through computer-readable storage media and computer equipment. The specific steps include obtaining pictures, using differentiable binary algorithm to obtain license plate and inkjet position information, using regression algorithm to mark and supplement position information, deleting overlapping information, and finally coding the license plate and inkjet position.
Effective detection and coding of small license plates and inkjet codes is achieved, manual participation is avoided, the fairness and safety of the inspection is improved, and the accuracy of vehicle sticker quality evaluation is ensured.
Smart Images

Figure CN114549272B_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of image processing, and in particular, relates to a method for coding a license plate and a spray code in a vehicle sticker, a computer-readable storage medium, and a computer device. Background Art
[0002] Since the company is vigorously promoting the car sticker service in various cities, the quality requirements for the car stickers are relatively high. The company will conduct spot checks on the car stickers by city every day to check the quality of the car stickers, and it is also directly linked to the performance of the city. In order to ensure the fairness of the review, a third-party company is introduced to conduct quality inspections; after the introduction of third-party quality inspections, data security is involved. In order to ensure the security of driver information on our company's platform and the fairness of reviewers, it is necessary to encode the key information of the provided image data. However, it is currently mainly carried out in the form of manual coding. The most common vehicle-related inspections are license plate inspections, and are limited to larger target license plates. It is not suitable for smaller license plates, and there is no inspection of spray codes. Summary of the invention
[0003] The purpose of this application is to provide a method for coding the license plate and the inkjet code in the car sticker, a computer-readable storage medium and a computer device, aiming to solve the problem that the current coding is not suitable for the detection of small license plates and there is no detection of the inkjet code.
[0004] In a first aspect, the present application provides a method for coding a license plate and a code in a car sticker, comprising:
[0005] Get the picture containing the car sticker;
[0006] According to a differentiable binarization algorithm, the position information of the license plate and the inkjet code is obtained from the picture containing the car sticker, and the picture with the position information of the license plate and the inkjet code marked is output;
[0007] According to the regression algorithm, the position information of the license plate and the inkjet code is obtained from the picture on which the position information of the license plate and the inkjet code is annotated, and the picture on which the position information of the license plate and the inkjet code is annotated after the regression algorithm is applied is output;
[0008] The image with the position information of the license plate and the inkjet code is labeled after the regression algorithm, and the overlapping position information of the license plate and the inkjet code is deleted to obtain the final image with the position information of the license plate and the inkjet code labeled;
[0009] All license plates and inkjet codes in the final image with the position information of the license plates and inkjet codes marked are coded.
[0010] Further, the method of obtaining the position information of the license plate and the inkjet code from the picture containing the car sticker according to the differentiable binarization algorithm and outputting the picture with the position information of the license plate and the inkjet code marked is specifically as follows: inputting the picture containing the car sticker into the model for training, marking the license plate and the inkjet code in the picture containing the car sticker, using the remaining part as the background, and segmenting the license plate and the inkjet code from the background to obtain the position information of the license plate and the inkjet code, marking a rectangular frame close to the license plate and the inkjet code according to the position information of the license plate and the inkjet code, and outputting the picture with the position information of the license plate and the inkjet code marked.
[0011] Furthermore, the step of obtaining the position information of the license plate and the inkjet code from the picture on which the position information of the license plate and the inkjet code is annotated according to the regression algorithm, and outputting the picture on which the position information of the license plate and the inkjet code is annotated after the regression algorithm is performed is specifically as follows: scaling the picture on which the position information of the license plate and the inkjet code is annotated, extracting visual features from the scaled picture based on the darknet53 network using the FPN algorithm and the PAN algorithm to obtain the position information of the license plate and the inkjet code, annotating a rectangular frame whose area is larger than the area of the license plate and the inkjet code according to the position information of the license plate and the inkjet code, and outputting the picture on which the position information of the license plate and the inkjet code is annotated after the regression algorithm is performed.
[0012] Furthermore, the image on which the position information of the license plate and the inkjet code is annotated after the regression algorithm deletes the overlapping position information of the license plate and the inkjet code to obtain the final image on which the position information of the license plate and the inkjet code is annotated. Specifically, from the image on which the position information of the license plate and the inkjet code is annotated after the regression algorithm, when the overlapping area of the rectangular frame that is closely attached to the license plate and the inkjet code and the rectangular frame whose annotated area is larger than the area of the license plate and the inkjet code is greater than or equal to half of the annotated rectangular frame, the overlapping rectangular frame whose annotated area is larger than the area of the license plate and the inkjet code is deleted.
[0013] Furthermore, after coding all the license plates and inkjet codes in the final picture with the position information of the license plates and inkjet codes marked, the method further includes: sending the coded picture to a quality inspection department for quality inspection of the vehicle stickers.
[0014] In a second aspect, the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for coding the license plate and the inkjet code in the vehicle sticker.
[0015] In a third aspect, the present application provides a computer device, comprising: one or more processors, a memory, and one or more computer programs, wherein the processor and the memory are connected via a bus, wherein the one or more computer programs are stored in the memory and are configured to be executed by the one or more processors, and when the processor executes the computer program, the steps of the method for coding the license plate and the inkjet code in the vehicle sticker are implemented.
[0016] In the present application, the position information of the license plate and the inkjet code is obtained according to the differentiable binarization algorithm; and then the position information of the license plate and the inkjet code is obtained according to the regression algorithm. The position information of the license plate and the inkjet code missed by the differentiable binarization algorithm can be supplemented, and the license plate and the inkjet code with a small target area can be coded. In the coding process, no human participation is required to avoid information leakage, thereby ensuring the fairness of the quality evaluation of the car sticker. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 It is a flow chart of a method for coding a license plate and a code in a car sticker provided in one embodiment of the present application.
[0018] Figure 2 It is a schematic diagram of a method for coding a license plate and a code in a car sticker provided in an embodiment of the present application.
[0019] Figure 3 It is a specific structural block diagram of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION
[0020] In order to make the purpose, technical solution and beneficial effects of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0021] In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.
[0022] See also Figure 1 The method for coding the license plate and the inkjet code in the car sticker provided in one embodiment of the present application includes the following steps: It should be noted that if there is substantially the same result, the method for coding the license plate and the inkjet code in the car sticker of the present application is not based on Figure 1 The process sequence shown is limited.
[0023] S101, obtaining a picture containing a car sticker;
[0024] S102, obtaining the position information of the license plate and the inkjet code from the picture containing the car sticker according to a differentiable binarization algorithm, and outputting a picture with the position information of the license plate and the inkjet code marked;
[0025] S103, according to the regression algorithm, obtaining the position information of the license plate and the inkjet code from the picture on which the position information of the license plate and the inkjet code is annotated, and outputting the picture on which the position information of the license plate and the inkjet code is annotated after the regression algorithm;
[0026] S104, deleting the overlapping license plate and inkjet position information from the image marked with the position information of the license plate and inkjet code after the regression algorithm, to obtain a final image marked with the position information of the license plate and inkjet code;
[0027] S105. Coding all license plates and inkjet codes in the final image with the position information of license plates and inkjet codes marked. In one embodiment of the present application, the method of obtaining the position information of license plates and inkjet codes from the image containing car stickers according to the differentiable binarization algorithm and outputting the image with the position information of license plates and inkjet codes marked is as follows: inputting the image containing car stickers into the model for training, marking the license plates and inkjet codes in the image containing car stickers, and using the remaining part as the background, and segmenting the license plates and inkjet codes from the background to obtain the position information of license plates and inkjet codes, marking a rectangular frame close to the license plates and inkjet codes according to the position information of license plates and inkjet codes, and outputting the image with the position information of license plates and inkjet codes marked.
[0028] In one embodiment of the present application, the position information of the license plate and the inkjet code is obtained from the picture on which the position information of the license plate and the inkjet code is marked according to the regression algorithm, and the picture on which the position information of the license plate and the inkjet code is marked after the regression algorithm is outputted. Specifically, the picture on which the position information of the license plate and the inkjet code is marked is scaled, and visual features of the scaled picture are extracted based on the darknet53 network using the FPN algorithm and the PAN algorithm to obtain the position information of the license plate and the inkjet code, and a rectangular frame with an area larger than the area of the license plate and the inkjet code is marked according to the position information of the license plate and the inkjet code, and the picture on which the position information of the license plate and the inkjet code is marked after the regression algorithm is outputted. In one embodiment of the present application, the image on which the position information of the license plate and the inkjet code is annotated after the regression algorithm is used to delete the overlapping position information of the license plate and the inkjet code to obtain the final image on which the position information of the license plate and the inkjet code is annotated. Specifically, from the image on which the position information of the license plate and the inkjet code is annotated after the regression algorithm, when the overlapping area of the rectangular frame that is closely attached to the license plate and the inkjet code and the rectangular frame whose annotated area is larger than the area of the license plate and the inkjet code is greater than or equal to half of the annotated rectangular frame, the overlapping rectangular frame whose annotated area is larger than the area of the license plate and the inkjet code is deleted.
[0029] In one embodiment of the present application, after coding all the license plates and inkjet codes in the final image with the position information of the license plates and inkjet codes marked, the method further includes: sending the coded image to the quality inspection department for quality inspection of the vehicle stickers.
[0030] For specific coding conditions, please refer to Figure 2 .
[0031] An embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the method for coding the license plate and the inkjet code in the vehicle sticker as provided in an embodiment of the present application are implemented.
[0032] Figure 3 A specific structural block diagram of a computer device provided in an embodiment of the present application is shown. A computer device 100 includes: one or more processors 101, a memory 102, and one or more computer programs, wherein the processor 101 and the memory 102 are connected via a bus, the one or more computer programs are stored in the memory 102, and are configured to be executed by the one or more processors 101, and when the processor 101 executes the computer program, the steps of the method for coding the license plate and the inkjet code in the vehicle sticker provided in an embodiment of the present application are implemented.
[0033] The computer device includes a server and a terminal, etc. The computer device may be a desktop computer, a mobile terminal or a vehicle-mounted device, and the mobile terminal includes at least one of a mobile phone, a tablet computer, a personal digital assistant or a wearable device, etc.
[0034] The embodiment of the present application obtains the position information of the license plate and the inkjet code based on the differentiable binarization algorithm; and then obtains the position information of the license plate and the inkjet code based on the regression algorithm, which can supplement the position information of the license plate and the inkjet code missed by the differentiable binarization algorithm, and encode the license plate and the inkjet code with a small target area. In addition, no human participation is required in the encoding process, which avoids information leakage and ensures the fairness of the quality evaluation of the vehicle sticker.
[0035] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.
[0036] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.
Claims
1. A method for coding a license plate and a code in a car sticker, characterized in that: include: Get the picture containing the car sticker; According to the differentiable binarization algorithm, the position information of the license plate and the inkjet code is obtained from the picture containing the car sticker, and the picture with the position information of the license plate and the inkjet code is output; specifically, the picture containing the car sticker is input into the model for training, the license plate and the inkjet code in the picture containing the car sticker are marked, and the remaining part is used as the background, and the license plate and the inkjet code are segmented from the background to obtain the position information of the license plate and the inkjet code, and a rectangular frame close to the license plate and the inkjet code is marked according to the position information of the license plate and the inkjet code, and the picture with the position information of the license plate and the inkjet code is output; According to the regression algorithm, the position information of the license plate and the inkjet code is obtained from the picture on which the position information of the license plate and the inkjet code is annotated, and the picture on which the position information of the license plate and the inkjet code is annotated after the regression algorithm is applied is output; specifically, the picture on which the position information of the license plate and the inkjet code is annotated is scaled, and visual features are extracted from the scaled picture based on the darknet53 network using the FPN algorithm and the PAN algorithm to obtain the position information of the license plate and the inkjet code, and a rectangular frame with an area larger than the area of the license plate and the inkjet code is annotated according to the position information of the license plate and the inkjet code, and the picture on which the position information of the license plate and the inkjet code is annotated after the regression algorithm is applied is output; The overlapping position information of the license plate and the inkjet code is deleted from the image marked with the position information of the license plate and the inkjet code after the regression algorithm, so as to obtain the final image marked with the position information of the license plate and the inkjet code; specifically, from the image marked with the position information of the license plate and the inkjet code after the regression algorithm, when the overlapping area of the rectangular frame marked close to the license plate and the inkjet code and the rectangular frame whose marked area is larger than the area of the license plate and the inkjet code is greater than or equal to half of the marked rectangular frame, the overlapping rectangular frame whose marked area is larger than the area of the license plate and the inkjet code is deleted; All license plates and inkjet codes in the final image with the position information of the license plates and inkjet codes marked are coded.
2. The method according to claim 1, characterized in that After coding all the license plates and inkjet codes in the final picture with the position information of the license plates and inkjet codes marked, the method further includes: sending the coded picture to a quality inspection department for quality inspection of the vehicle stickers.
3. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, the steps of the method for coding the license plate and the inkjet code in the vehicle sticker as claimed in claim 1 or 2 are implemented.
4. A computer device comprising: One or more processors, a memory and one or more computer programs, the processor and the memory are connected via a bus, wherein the one or more computer programs are stored in the memory and are configured to be executed by the one or more processors, characterized in that when the processor executes the computer program, the steps of the method for coding the license plate and the inkjet code in the vehicle sticker as described in claim 1 or 2 are implemented.
Citation Information
Patent Citations
Document title recognition method and system based on deep learning, terminal and medium
CN112766246A