License plate recognition method and device, electronic equipment and storage medium

Abstract

The invention provides a license plate recognition method and device, electronic equipment and a storage medium. Relates to the technical field of computers, in particular to the fields of Internet of Vehicles, Internet of Things, deep learning, artificial intelligence, smart cities and the like. The method comprises the following steps: determining the number of license plate characters of a license plate and license plate character position information of each license plate character according to the license plate type of the license plate in an image; performing character recognition on the license plate area in the image to obtain a predicted character sequence and predicted character position information of each predicted character in the predicted character sequence; in response to determining that the predicted character number of the predicted character sequence is smaller than the license plate character number, determining missing character position information; performing character recognition on a missing character region represented by the missing character position information to obtain a missing character recognition result; and determining a license plate character recognition result of the license plate according to the predicted character sequence, the missing character recognition result and the missing character position information.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to the fields of Internet of Vehicles, Internet of Things, artificial intelligence, smart cities, and more specifically to a license plate recognition method, device, electronic device, and storage medium. Background Art

[0002] Automatic and accurate recognition of vehicle license plates has many important applications. For example, it can be used to analyze the working status of vehicles, thereby visually displaying the occupancy of resources such as vehicles and platforms, and can also provide important support for intelligent resource scheduling. However, in logistics transfer centers, especially when vehicles are loading and unloading at platforms, the vehicle license plate numbers are still mainly recorded manually.

[0003] In the process of realizing the concept disclosed in the present invention, the inventors found that there are at least the following problems in the related technology: the manual recording of license plate numbers not only wastes valuable human resources, but also cannot provide support for subsequent visual display and resource scheduling due to the lack of digitization, thus seriously affecting the operating efficiency of the entire park. Summary of the invention

[0004] In view of this, the present disclosure provides a license plate recognition method, device, electronic device and storage medium.

[0005] One aspect of the present disclosure provides a license plate recognition method, comprising: determining the number of license plate characters and license plate character position information of each license plate character according to the license plate type of the license plate in an image; performing character recognition on the license plate area in the image to obtain a predicted character sequence and predicted character position information of each predicted character in the predicted character sequence; determining missing character position information in response to determining that the predicted number of characters in the predicted character sequence is less than the number of license plate characters; performing character recognition on the missing character area represented by the missing character position information to obtain a missing character recognition result; and determining the license plate character recognition result of the license plate according to the predicted character sequence, the missing character recognition result and the missing character position information.

[0006] Another aspect of the present disclosure provides a license plate recognition device, including: a license plate character information determination module, used to determine the number of license plate characters of the license plate and the license plate character position information of each license plate character according to the license plate type of the license plate in the image; a predicted character information determination module, used to perform character recognition on the license plate area in the image to obtain a predicted character sequence and predicted character position information of each predicted character in the predicted character sequence; a missing character position information determination module, used to determine missing character position information in response to determining that the predicted number of characters of the predicted character sequence is less than the number of license plate characters; a missing character recognition module, used to perform character recognition on the missing character area represented by the missing character position information to obtain a missing character recognition result; and a license plate character recognition result determination module, used to determine the license plate character recognition result of the license plate according to the predicted character sequence, the missing character recognition result and the missing character position information.

[0007] Another aspect of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the license plate recognition method described in the present disclosure.

[0008] Another aspect of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, which are used to implement the license plate recognition method described in the present disclosure when executed.

[0009] Another aspect of the present disclosure provides a computer program product, which includes computer executable instructions, and when the instructions are executed, they are used to implement the license plate recognition method described in the present disclosure.

[0010] According to the embodiments of the present disclosure, because the number of license plate characters and the license plate character position information of each license plate character are determined according to the license plate type of the license plate in the image; character recognition is performed on the license plate area in the image to obtain a predicted character sequence and predicted character position information of each predicted character in the predicted character sequence; in response to determining that the predicted number of characters of the predicted character sequence is less than the number of license plate characters, the missing character position information is determined; character recognition is performed on the missing character area represented by the missing character position information to obtain a missing character recognition result; and a technical means is used to determine the license plate character recognition result of the license plate according to the predicted character sequence, the missing character recognition result and the missing character position information. This can utilize prior knowledge such as the license plate type to automatically realize character recognition of the license plate area and make targeted modifications to the predicted unreasonable license plates. For example, the missing characters can be compensatory completed, which at least partially overcomes the technical problems of manually recording license plate numbers, such as wasting human resources and reducing operating efficiency, thereby achieving the technical effects of improving the accuracy of license plate recognition results and improving operating efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0012] Figure 1 An exemplary system architecture to which a license plate recognition method can be applied according to an embodiment of the present disclosure is schematically shown;

[0013] Figure 2 A flowchart of a license plate recognition method according to an embodiment of the present disclosure is schematically shown;

[0014] Figure 3A A schematic diagram of a single-line seven-character type license plate according to an embodiment of the present disclosure is schematically shown;

[0015] Figure 3B A schematic diagram of a single-row eight-character type license plate according to an embodiment of the present disclosure is schematically shown;

[0016] Figure 3C A schematic diagram of a double-row seven-character type license plate according to an embodiment of the present disclosure is schematically shown;

[0017] Figure 3D A schematic diagram of a double-row eight-character type license plate according to an embodiment of the present disclosure is schematically shown;

[0018] Figure 4 A schematic diagram of detecting license plate characters based on a model according to an embodiment of the present disclosure is schematically shown;

[0019] Figure 5 A block diagram schematically shows a license plate recognition device according to an embodiment of the present disclosure; and

[0020] Figure 6 A block diagram of an electronic device suitable for implementing a license plate recognition method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0021] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0022] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0023] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0024] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0025] It should be noted that the collection, update, analysis, processing, use, transmission, provision, disclosure, storage and other aspects of the data involved in the technical solution of this disclosure (for example, including but not limited to user personal information) are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures have been taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information, network security and national security.

[0026] In the embodiments of the present disclosure, the user's authorization or consent is obtained before obtaining or collecting the user's personal information.

[0027] License plate recognition is an important task that needs to be solved in logistics parks. Most solutions for automatic license plate recognition are based on computer vision technology. Generally speaking, this type of method first locates the regional position of the license plate in the entire image. License plate recognition is to identify the content of each character in the license plate area detected in the first step, and then combine it into the correct license plate number through a series of logical judgments. The automatic license plate recognition systems deployed in the industry, such as license plate recognition systems in parking lots or communities, have low camera positions and mostly recognize the front license plate of the vehicle. The angle is fixed, and the vehicle is usually required to stop in front of the gate, which is less difficult. According to different technical routes, the industry's vehicle recognition systems can be roughly divided into two categories: first recognize a single character and then synthesize the license plate number, and recognize the license plate number of the entire license plate based on a sequence model.

[0028] In the process of implementing the concept of the present disclosure, the inventors found that most solutions that first recognize individual characters and then synthesize them use a target detection model based on deep learning, with each character as a target, and then post-processing the detected characters to obtain a character string. This type of solution has no correction mechanism for outputting unreasonable license plate results, and the accuracy needs to be further improved. The solution that uses a sequence model to recognize the entire license plate does not segment and recognize individual characters, but uses the entire license plate as input and outputs the final license plate character sequence. This type of model can only recognize single-layer license plates and cannot recognize double-row rear license plates in the scenario of reversing on the platform for loading and unloading operations.

[0029] In addition, the camera position in the platform scenes in the logistics park is relatively high. Since the working state of the vehicle is mostly reverse loading and unloading, the rear license plate of the vehicle is mostly identified, and it cannot interfere with the normal work on site. Therefore, the vehicle cannot stop to cooperate with the license plate recognition. When the camera is located at a high position, the license plate captured will be smaller and severely blocked. The rear license plate will appear in two formats: single-line and double-line. This requires the design of reasonable single-double-line logical judgment and double-line format character combination. Failure to cooperate with parking, especially when leaving the platform at a high speed, will cause the license plate to be blurred and double-imaged, making detection difficult, thereby reducing the accuracy of license plate recognition.

[0030] The embodiments of the present disclosure provide a license plate recognition method, device, electronic device and storage medium. The method includes: determining the number of license plate characters and license plate character position information of each license plate character according to the license plate type in the image; performing character recognition on the license plate area in the image to obtain a predicted character sequence and predicted character position information of each predicted character in the predicted character sequence; determining missing character position information in response to determining that the predicted character number of the predicted character sequence is less than the number of license plate characters; performing character recognition on the missing character area represented by the missing character position information to obtain a missing character recognition result; and determining the license plate character recognition result of the license plate according to the predicted character sequence, the missing character recognition result and the missing character position information.

[0031] Figure 1 The exemplary system architecture 100 to which the license plate recognition method according to the embodiment of the present disclosure can be applied is schematically shown. It should be noted that: Figure 1 What is shown is merely an example of a system architecture to which the embodiments of the present disclosure can be applied, in order to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios.

[0032] like Figure 1 As shown, the system architecture 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired and / or wireless communication links, etc.

[0033] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, and / or social platform software, etc. (only for example).

[0034] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0035] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0036] It should be noted that the license plate recognition method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the license plate recognition device provided in the embodiment of the present disclosure can generally be set in the server 105. The license plate recognition method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Correspondingly, the license plate recognition device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Alternatively, the license plate recognition method provided in the embodiment of the present disclosure can also be executed by the first terminal device 101, the second terminal device 102 or the third terminal device 103, or can also be executed by other terminal devices that are different from the first terminal device 101, the second terminal device 102 or the third terminal device 103. Correspondingly, the license plate recognition device provided in the embodiment of the present disclosure may also be arranged in the first terminal device 101, the second terminal device 102 or the third terminal device 103, or in other terminal devices different from the first terminal device 101, the second terminal device 102 or the third terminal device 103.

[0037] For example, the image may be originally stored in any one of the first terminal device 101, the second terminal device 102 or the third terminal device 103 (for example, the first terminal device 101, but not limited thereto), or stored on an external storage device and imported into the first terminal device 101. Then, the first terminal device 101 may locally execute the license plate recognition method provided by the embodiment of the present disclosure, or send the image to other terminal devices, servers, or server clusters, and the other terminal devices, servers, or server clusters that receive the image may execute the license plate recognition method provided by the embodiment of the present disclosure.

[0038] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0039] Figure 2A flowchart of a license plate recognition method according to an embodiment of the present disclosure is schematically shown.

[0040] like Figure 2 As shown, the method includes operations S201 to S205.

[0041] In operation S201, the number of license plate characters of the license plate and license plate character position information of each license plate character are determined according to the license plate type of the license plate in the image.

[0042] According to an embodiment of the present disclosure, the image may be an image or video frame captured by various image acquisition devices or video acquisition devices such as cameras for the license plate. Based on the prior knowledge contained in the license plate, the license plate may be classified, for example, according to the number of characters and the arrangement of the characters. In this embodiment, the types of license plates obtained by classifying the license plates may include, for example: single-line seven characters, single-line eight characters, double-line seven characters, double-line eight characters, etc., and may not be limited thereto.

[0043] Figure 3A A schematic diagram of a single-row seven-character type license plate according to an embodiment of the present disclosure is schematically shown.

[0044] Figure 3B A schematic diagram of a single-row eight-character type license plate according to an embodiment of the present disclosure is schematically shown.

[0045] Figure 3C A schematic diagram of a double-row seven-character type license plate according to an embodiment of the present disclosure is schematically shown.

[0046] Figure 3D A schematic diagram of a double-row eight-character type license plate according to an embodiment of the present disclosure is schematically shown.

[0047] like Figure 3A to Figure 3D As shown, the first digit of each type of license plate is a character representing the abbreviation of a province. There are 32 characters representing the abbreviation of provinces to choose from, and the remaining digits can be any one of the 10 Arabic numerals and 24 English letters excluding I and O.

[0048] According to an embodiment of the present disclosure, images or video frames including different types of license plates can be collected as training data according to the license plate type, and a deep learning-based target detection technology can be used to train a deep learning-based license plate detection model using images or video frames including multiple types of license plate information as input. Through the output of the license plate detection model, the position of the license plate area in the image, the number of characters in the license plate (7 bits or 8 bits), and the arrangement of the characters in the license plate can be obtained, that is, the license plate type can be determined. At the same time, the corresponding position of each character in the license plate can be determined according to the different types of license plates. The trained license plate detection model is able to detect the license plate type, and classify the license plate and recognize the characters according to the different types of license plates.

[0049] It should be noted that the above-mentioned training of the license plate detection model to identify the license plate type and then determine the number of license plate characters and the license plate character position information is only an exemplary embodiment and is not limited to this. For example, in some embodiments, the license plate information in the image can also be matched with the known license plate type to determine the license plate type of the license plate in the image. Then, according to the license plate type, the number of license plate characters and the arrangement of the license plate characters are determined, and according to the arrangement of the license plate characters, the license plate area in the image is divided into regions to determine the license plate character position information of each license plate character.

[0050] In operation S202, character recognition is performed on the license plate area in the image to obtain a predicted character sequence and predicted character position information of each predicted character in the predicted character sequence.

[0051] According to an embodiment of the present disclosure, the license plate area can be first determined based on the position of the license plate in the image. Then, a character recognition model can be trained to identify the license plate area by classifying each character as a target to be detected, output the detected license plate characters, and design a series of post-processing logic to combine individual characters into a more reliable string to obtain a predicted character sequence, which can characterize the detected license plate results. In this process, the character recognition model can also output the position information of each detected license plate character, that is, obtain the predicted character position information. The post-processing logic may include an algorithm for judging single and double rows, an algorithm for removing abnormal characters, etc., and may not be limited to this.

[0052] According to an embodiment of the present disclosure, character recognition can be limited to detecting 66 types of characters including 10 Arabic numerals, 24 English letters (excluding I and O) and 32 characters representing provincial abbreviations, which can help improve the detection efficiency and detection accuracy of character recognition. The above-mentioned abnormal characters can represent other characters that do not belong to the 66 types of character positions.

[0053] In operation S203, in response to determining that the number of predicted characters of the predicted character sequence is less than the number of license plate characters, missing character position information is determined.

[0054] According to the embodiments of the present disclosure, the detected license plate result may be incomplete due to factors such as environmental occlusion and incomplete character recognition. In this embodiment, for example, by judging whether the predicted number of characters of the predicted character sequence is consistent with the number of license plate characters determined according to the license plate type, it is possible to determine whether there are missing characters in the predicted character sequence and determine the position information of the missing characters. For example, when it is determined that the predicted number of characters is less than the number of license plate characters, it indicates that there are undetected characters in the predicted character sequence, that is, missing characters.

[0055] According to an embodiment of the present disclosure, when it is determined that there are missing characters, the position of the missing characters can be located according to the arrangement order of the characters in the single-row and double-row license plates. For example, the position information of the missing characters can be determined by matching the position information occupied by the predicted characters in the predicted character sequence with the license plate character position information.

[0056] According to an embodiment of the present disclosure, the missing character position information can also be determined based on the character category of the predicted character sequence, when it is possible to determine that a certain category of characters is missing and to determine the position that the missing characters of the category should occupy based on the arrangement of the characters of each category represented by the license plate type.

[0057] It should be noted that the above-mentioned method for determining the missing character position information is only an exemplary embodiment, and the art may also include other embodiments to assist in determining the missing character position information, which may not be limited to the above-mentioned.

[0058] In operation S204, character recognition is performed on the missing character region represented by the missing character position information to obtain a missing character recognition result.

[0059] According to the embodiments of the present disclosure, the missing character position information can be represented by a pixel position information, or can be represented by information including the pixel position information and the width, height and other information determined with the pixel position information as the center (or any point in the area). After determining the missing character position information, the missing character area can be determined based on the missing character position information.

[0060] According to an embodiment of the present disclosure, data of 66 types of characters may be first collected to train a classification model. The classification model may be a convolutional neural network or a deep network of a Transformer, which is not limited here. Then, the missing character region may be input into the trained classification model, and character recognition may be performed on the missing character region separately to obtain a missing character recognition result.

[0061] In operation S205, a license plate character recognition result of the license plate is determined according to the predicted character sequence, the missing character recognition result and the missing character position information.

[0062] According to an embodiment of the present disclosure, the operation S205 may include: inserting the missing character sequence into the corresponding position of the predicted character sequence according to the missing character position information to obtain a license plate character recognition result.

[0063] It should be noted that the specific implementation process of the above operation S205 can be determined by manual decision-making or by computer automation, which is not limited here. In the above operation S203, when it is determined that the number of predicted characters is consistent with the number of license plate characters, the predicted character sequence can be determined as the license plate character recognition result of the license plate in combination with post-processing logic.

[0064] Figure 4 A schematic diagram of detecting license plate characters based on a model according to an embodiment of the present disclosure is schematically shown.

[0065] like Figure 4 As shown, the model 400 may include a license plate detection model 410, a character recognition model 420, and a classification model 430. The image or video frame 401 containing the license plate is input into the license plate detection model 410, and the license plate position information 411 and the license plate type 412 can be output. According to the license plate position information 411, a license plate area 413 can be cut out from the image or video frame 401. The license plate area 413 is input into the character recognition model 420, and a predicted character sequence 421 can be obtained. In the case where it is determined that the number of predicted characters according to the predicted character sequence 421 is less than the number of license plate characters determined according to the license plate type 412, missing character position information 402 can be obtained. According to the missing character position information 402, a missing character area 403 can be cut out from the license plate area 413. The missing character area 403 is input into the classification model 430, and a missing character recognition result 431 can be obtained. By obtaining the predicted character sequence 421, the missing character recognition result 431 and the missing character position information 402, the missing characters can be filled in according to the positional relationship between the characters to obtain the license plate character recognition result 404 of the license plate.

[0066] Through the above-mentioned embodiments of the present disclosure, when the number of characters output by the license plate recognition model does not reach the number of characters in the real license plate, the undetected characters can be located through the prior knowledge of the license plate, and then the characters can be classified to fill in the missing characters, thereby achieving compensatory completion of the license plate prediction output by the model. The technical problems of wasting human resources and reducing operating efficiency by manually recording license plate numbers are at least partially overcome, thereby achieving technical effects such as improving the accuracy of license plate recognition results and improving operating efficiency.

[0067] In conjunction with specific embodiments, Figure 2 The method shown is further explained.

[0068] According to an embodiment of the present disclosure, the above operation S203 may include: determining missing character position information according to character information of the predicted character sequence.

[0069] According to the embodiments of the present disclosure, the types of license plate characters at corresponding positions in various types of license plates can be determined according to the license plate type. For example, it can be determined that the leftmost character of a single-row license plate should be a character representing the abbreviation of a province, the leftmost character of the upper row of a double-row license plate should be a character representing the abbreviation of a province, and the second character of various types of license plates should be a letter, etc.

[0070] According to an embodiment of the present disclosure, the above-mentioned determination of the missing character position information based on the character information of the predicted character sequence may also include: in response to determining that the predicted character sequence is missing the province character representing the province, determining the missing character position information based on the province character area position information representing the province character in the license plate character position information.

[0071] For example, when it is determined that the first position of the detected license plate character is not a character representing a province abbreviation, the information of the position that the character representing the province abbreviation should occupy can be determined as missing character position information according to the license plate type. For example, the position information occupied by the character miniature on the leftmost side of a single-row license plate and the position information occupied by the character miniature on the leftmost side of the upper row of a double-row license plate can be determined as missing character position information.

[0072] According to an embodiment of the present disclosure, the above-mentioned determination of missing character position information based on character information of the predicted character sequence may also include: in response to determining that the predicted character sequence is missing English letters, determining the missing character position information based on English letter area position information representing English letters in the license plate character position information.

[0073] For example, when it is determined that the second digit of the detected license plate character is not a letter, the license plate character position information occupied by the second license plate character may be determined as missing character position information.

[0074] According to an embodiment of the present disclosure, the above operation S203 may further include: determining missing character position information according to at least one of the license plate character position information, predicted character position information and the license plate type.

[0075] According to an embodiment of the present disclosure, the predicted character sequence may include N predicted characters, where N may be a positive integer. When the license plate type is a single-line seven characters or a double-line seven characters, and N is an integer less than 7, it may be determined that the predicted character sequence is incomplete and there are missing characters. When the license plate type is a single-line eight characters or a double-line eight characters, and N is an integer less than 8, it may also be determined that the predicted character sequence is incomplete and there are missing characters.

[0076] According to an embodiment of the present disclosure, the above-mentioned determining the missing character position information according to at least one of the license plate character position information, the predicted character position information and the license plate type may also include: obtaining the Nth predicted character position information of the Nth predicted character according to the predicted character position information. Obtaining the second target license plate character position information of the last license plate character according to the license plate character position information. In response to determining that the overlap between the Nth predicted character and the last license plate character is less than a preset threshold according to the Nth predicted character position information and the second target license plate character position information, the second target license plate character position information is determined as the missing character position information.

[0077] According to an embodiment of the present disclosure, the above method can also be applied to an embodiment of matching any predicted character and license plate character to determine missing character position information. For example, the license plate character position information and the predicted character position information can be matched, and when it is determined that the distance between the position of a certain license plate character represented by the license plate character position information and the position of a certain predicted character represented by the predicted character position information is less than a distance threshold, it can be determined that the license plate character position information of the license plate character matches the predicted character position information of the predicted character. Otherwise, the two do not match. In the case of determining that there is license plate character position information that is not matched to the predicted character position information, the license plate character position information that is not matched to the predicted character position information can be determined as missing character position information.

[0078] According to an embodiment of the present disclosure, the missing character position information can also be determined based on the license plate character position information, the predicted character position information and the license plate type. For example, the arrangement of the license plate characters in the license plate area and the required spacing between adjacent license plate characters can be first determined based on the license plate type. Then, the missing character position information can be determined based on the predicted character position information and the spacing.

[0079] It should be noted that the above method of determining the missing character position information based on at least one of the license plate character position information, the predicted character position information and the license plate type is only an exemplary embodiment and is not limited thereto. The art can also determine the missing character position information in combination with other license plate attribute information and other methods, which are not limited here.

[0080] According to an embodiment of the present disclosure, the above-mentioned determination of missing character position information based on license plate character position information, predicted character position information and license plate type may include: determining the average license plate character width of the license plate characters based on the number of license plate characters and the license plate character position information. Determining the license plate character spacing threshold between adjacent license plate characters based on the average license plate character width and the license plate type. In response to determining that the distance between the nth predicted character and the n+1th predicted character is greater than the license plate character spacing threshold, obtaining the nth predicted character position information of the nth predicted character and the n+1th predicted character position information of the n+1th predicted character, wherein n is an integer greater than or equal to 1 and less than N. The first target license plate character position information located between the nth predicted character position information and the n+1th predicted character position information is determined as the missing character position information.

[0081] For example, in the case where it is determined that there is a missing character based on the number of license plate characters adjusted by the license plate character sequence and the number of predicted characters represented by the predicted character sequence, the average width w of the license plate characters can be calculated first. Then, for example, the license plate character spacing threshold can be determined to be 0.6w based on the average width w. Thereafter, the following steps can be performed: in the case where it is determined that the distance between the nth predicted character and the n+1th predicted character is greater than 0.6w, the first target license plate character position information located between the nth predicted character position information of the nth predicted character and the n+1th predicted character position information of the n+1th predicted character is determined as the missing character position information.

[0082] According to an embodiment of the present disclosure, the above-mentioned determination of the license plate character spacing threshold between adjacent license plate characters based on the average width of the license plate characters and the license plate type may include: in response to determining that the license plate type is a single-row license plate, determining the license plate character spacing threshold between first adjacent license plate characters as a first spacing threshold, wherein the first adjacent license plate characters are the second license plate characters and the third license plate characters. Determining the license plate character spacing threshold between second adjacent license plate characters as a second spacing threshold, wherein the second adjacent license plate characters are other adjacent license plate characters in the license plate characters of the single-row license plate except the first adjacent license plate characters, and the first spacing threshold is greater than the second spacing threshold.

[0083] According to the embodiments of the present disclosure, based on the prior knowledge of the license plate, it can be determined that the interval between the second license plate character and the third license plate character in a single-line license plate is larger than the interval between other adjacent license plate characters. Therefore, a larger license plate character interval threshold, such as 1.1w, can be determined based on the average width w as the first interval threshold. And 0.6w can still be determined as the second interval threshold.

[0084] It should be noted that the above method of determining the license plate character spacing threshold, the first spacing threshold and the second spacing threshold is only an exemplary embodiment. In actual business, the corresponding thresholds can also be adaptively adjusted according to business needs, which is not limited to the above.

[0085] Taking the case where there is a missing character in the result of the single-line seven-character license plate recognition as an example, the average width w of the license plate characters can be calculated first. Then, the following judgment process can be performed in sequence: If the first character is not a province, the first position is lost. There are two ways to judge whether the second position is missing. If the second character is not a letter, the second position is lost. If the distance between the first character and the second character (the left boundary of the second character minus the right boundary of the first character) is greater than or equal to 0.6 times w, the second position is lost. If the distance between the second character and the third character is greater than or equal to 1.1w, the third position is lost (the reason why it is not 0.6W is that the spacing between the second character and the third character is greater than the spacing between other characters). If the distance between the third character and the fourth character is greater than or equal to 0.6 times w, the fourth position is lost. If the distance between the fourth character and the fifth character is greater than or equal to 0.6 times w, the fifth position is lost. If the distance between the fifth character and the sixth character is greater than or equal to 0.6 times w, the sixth position is lost. At this point, the missing bit is the last one.

[0086] It should be noted that the above only exemplifies an embodiment in which the detection result of a single-line seven-character type license plate is missing one digit. Those skilled in the art should understand that the above method can also be applied to the case in which the detection result of a single-line seven-character type license plate is missing multiple digits, as well as the case in which various types of license plate recognition results such as single-line eight characters, double-line seven characters, and double-line eight characters are missing one or more digits. In each case, only adaptive adjustment is required based on the actual situation, which will not be elaborated here.

[0087] According to an embodiment of the present disclosure, determining the missing character position information according to the predicted character position information and the license plate type may include: in response to determining that the license plate type is a double-row license plate, determining the center position information of each predicted character in the first row of predicted characters and at least one of the left boundary position information and the right boundary position information of each predicted character in the second row of predicted characters according to the predicted character position information. Determine the missing character position information according to at least one of the left boundary position information and the right boundary position information and the center position information.

[0088] Taking the case where there is a missing character in the result of the double-line seven-character license plate recognition as an example, first, according to the prior information of the license plate, it can be determined that the license plate characters of the double-line seven-character license plate are arranged as follows: 2 in the upper row and 5 in the lower row. The judgment method for missing a character in the upper row may include: judging whether there is a province, if there is no province, the missing character is the province. Judging according to the horizontal center position information of the predicted character frame obtained in the upper row, comparing it with the center position information of the third character in the lower row, if it is greater than, the first position is lost, and if it is less than, the second position is lost. When a character is missing in the lower row, the following judgment process is performed in sequence: if the center position information of the first character in the lower row is greater than the left boundary of the province, the missing character is the first position in the lower row. If the center position information of the second character in the lower row is greater than the right boundary of the province, the missing character is the second position in the lower row. If the center position information of the third character in the lower row is greater than the left boundary of the second character in the upper row, the missing character is the third position in the lower row. If the center position information of the fourth character in the lower row is greater than the right boundary of the second character in the upper row, the missing character is the fourth position in the lower row. At this point, the missing character is the fifth position in the lower row.

[0089] It should be noted that after determining which character is missing based on the above method, the missing character position information can be determined in combination with the license plate character position information.

[0090] It should be noted that the above only exemplifies an embodiment in which the detection result of a double-line seven-character type license plate is missing one digit. Those skilled in the art should understand that the above method can also be applied to the case in which the detection result of a double-line seven-character type license plate is missing multiple digits, and the case in which the recognition result of a double-line eight-character type license plate is missing one or more digits. It only needs to be adjusted accordingly according to the actual situation, which will not be elaborated here.

[0091] Through the above-mentioned embodiments of the present disclosure, a method for identifying license plates by utilizing prior knowledge of license plates is proposed, which can be applied in places equipped with cameras, such as platforms in logistics parks. By using video equipment that monitors the platform to collect videos of vehicles entering and leaving the platform, and continuously extracting video frames as input for license plate recognition, a completion strategy can be implemented when there are missing characters in the prediction results. Based on this strategy, a fine-grained license plate recognition method can be implemented, by outputting the number of license plate characters and single and double line type information, and using license plate prior information to locate missing characters, combined with the use of a classification model to predict and complete the missing characters, the efficiency and accuracy of license plate recognition can be effectively improved.

[0092] Figure 5 A block diagram of a license plate recognition device according to an embodiment of the present disclosure is schematically shown.

[0093] like Figure 5As shown, the license plate recognition device 500 includes a license plate character information determination module 510, a predicted character information determination module 520, a missing character position information determination module 530, a missing character recognition module 540 and a license plate character recognition result determination module 550.

[0094] The license plate character information determination module 510 is used to determine the number of license plate characters and the license plate character position information of each license plate character according to the license plate type of the license plate in the image.

[0095] The predicted character information determination module 520 is used to perform character recognition on the license plate area in the image to obtain a predicted character sequence and predicted character position information of each predicted character in the predicted character sequence.

[0096] The missing character position information determination module 530 is used to determine the missing character position information in response to determining that the number of predicted characters in the predicted character sequence is less than the number of license plate characters.

[0097] The missing character recognition module 540 is used to perform character recognition on the missing character region represented by the missing character position information to obtain a missing character recognition result.

[0098] The license plate character recognition result determination module 550 is used to determine the license plate character recognition result of the license plate according to the predicted character sequence, the missing character recognition result and the missing character position information.

[0099] According to an embodiment of the present disclosure, the missing character position information determining module includes a first missing character position information determining submodule.

[0100] The first missing character position information determining submodule is used to determine the missing character position information according to the character information of the predicted character sequence.

[0101] According to an embodiment of the present disclosure, the first missing character position information determining submodule includes a first missing character position information determining unit.

[0102] The first missing character position information determining unit is used to determine the missing character position information according to the province character area position information representing the province character in the license plate character position information in response to determining that the predicted character sequence is missing the province character representing the province.

[0103] According to an embodiment of the present disclosure, the first missing character position information determining submodule includes a second missing character position information determining unit.

[0104] The second missing character position information determining unit is used to determine the missing character position information according to the English letter area position information representing the English letters in the license plate character position information in response to determining that the predicted character sequence is missing English letters.

[0105] According to an embodiment of the present disclosure, the missing character position information determining module includes a second missing character position information determining submodule.

[0106] The second missing character position information determining submodule is used to determine the missing character position information according to at least one of the license plate character position information, the predicted character position information and the license plate type.

[0107] According to an embodiment of the present disclosure, the predicted character sequence includes N predicted characters, where N is a positive integer. The second missing character position information determination submodule includes a license plate character average width determination unit, a license plate character spacing threshold determination unit, a predicted character position information determination unit, and a third missing character position information determination unit.

[0108] The license plate character average width determination unit is used to determine the license plate character average width of the license plate characters according to the number of license plate characters and the license plate character position information.

[0109] The license plate character spacing threshold determination unit is used to determine the license plate character spacing threshold between adjacent license plate characters according to the average width of the license plate characters and the license plate type.

[0110] A predicted character position information determination unit is used to obtain the nth predicted character position information of the nth predicted character and the n+1th predicted character position information of the n+1th predicted character in response to determining that the distance between the nth predicted character and the n+1th predicted character is greater than the license plate character spacing threshold, wherein n is an integer greater than or equal to 1 and less than N.

[0111] The third missing character position information determining unit is used to determine the first target license plate character position information located between the nth predicted character position information and the (n+1)th predicted character position information as the missing character position information.

[0112] According to an embodiment of the present disclosure, the license plate character interval threshold determination unit includes a first interval threshold determination subunit and a second interval threshold determination subunit.

[0113] The first interval threshold determination subunit is used to determine the license plate character interval threshold between first adjacent license plate characters as the first interval threshold in response to determining that the license plate type is a single-row license plate, wherein the first adjacent license plate characters are the second license plate character and the third license plate character.

[0114] The second interval threshold determination subunit is used to determine the license plate character interval threshold between second adjacent license plate characters as the second interval threshold, wherein the second adjacent license plate characters are other adjacent license plate characters in the license plate characters of the single-row license plate except the first adjacent license plate characters, and the first interval threshold is greater than the second interval threshold.

[0115] According to an embodiment of the present disclosure, the second missing character position information determining submodule includes a predicted character information determining unit and a fourth missing character position information determining unit.

[0116] A predicted character information determination unit is used to determine, in response to determining that the license plate type is a double-row license plate, the center position information of each predicted character in the first row of predicted characters, and at least one of the left boundary position information and the right boundary position information of each predicted character in the second row of predicted characters based on the predicted character position information.

[0117] The fourth missing character position information determining unit is used to determine the missing character position information according to at least one of the left boundary position information and the right boundary position information and the center position information.

[0118] According to an embodiment of the present disclosure, the second missing character position information determining submodule includes an Nth predicted character position information acquiring unit, a second target license plate character position information acquiring unit and a fifth missing character position information determining unit.

[0119] The Nth predicted character position information obtaining unit is used to obtain the Nth predicted character position information of the Nth predicted character according to the predicted character position information.

[0120] The second target license plate character position information acquiring unit is used to acquire the second target license plate character position information of the last license plate character according to the license plate character position information.

[0121] The fifth missing character position information determination unit is used to determine, in response to the Nth predicted character position information and the second target license plate character position information, that the overlap between the Nth predicted character and the last license plate character is less than a preset threshold, and determine the second target license plate character position information as the missing character position information.

[0122] According to an embodiment of the present disclosure, the license plate character recognition result determination module includes a license plate character recognition result determination sub-module.

[0123] The license plate character recognition result determination submodule is used to insert the missing character sequence into the corresponding position of the predicted character sequence according to the missing character position information to obtain the license plate character recognition result.

[0124] According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits, or at least part of the functions of any one of them can be implemented in one module. According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits can be split into multiple modules for implementation. According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits can be at least partially implemented as hardware circuits, such as field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), systems on chips, systems on substrates, systems on packages, application specific integrated circuits (ASICs), or can be implemented by hardware or firmware in any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware, and firmware, or in any appropriate combination of any of them. Alternatively, according to the embodiments of the present invention, one or more of the modules, submodules, units, and subunits can be at least partially implemented as computer program modules, and when the computer program modules are run, the corresponding functions can be performed.

[0125] For example, any multiple of the license plate character information determination module 510, the predicted character information determination module 520, the missing character position information determination module 530, the missing character recognition module 540, and the license plate character recognition result determination module 550 can be combined in one module / submodule / unit / subunit for implementation, or any one of the modules / submodules / units / subunits can be split into multiple modules / submodules / units / subunits. Alternatively, at least part of the functions of one or more of these modules / submodules / units / subunits can be combined with at least part of the functions of other modules / submodules / units / subunits and implemented in one module / submodule / unit / subunit. According to an embodiment of the present disclosure, at least one of the license plate character information determination module 510, the predicted character information determination module 520, the missing character position information determination module 530, the missing character recognition module 540 and the license plate character recognition result determination module 550 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in an appropriate combination of any of them. Alternatively, at least one of the license plate character information determination module 510, the predicted character information determination module 520, the missing character position information determination module 530, the missing character recognition module 540 and the license plate character recognition result determination module 550 can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be performed.

[0126] It should be noted that the license plate recognition device part in the embodiments of the present disclosure corresponds to the license plate recognition method part in the embodiments of the present disclosure. The description of the license plate recognition device part specifically refers to the license plate recognition method part, which will not be repeated here.

[0127] Figure 6 A block diagram of an electronic device suitable for implementing a license plate recognition method according to an embodiment of the present disclosure is schematically shown. Figure 6 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0128] like Figure 6 As shown, the electronic device 600 according to an embodiment of the present disclosure includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include an onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0129] In RAM 603, various programs and data required for the operation of electronic device 600 are stored. Processor 601, ROM 602 and RAM 603 are connected to each other via bus 604. Processor 601 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 602 and / or RAM 603. It should be noted that the program can also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0130] According to an embodiment of the present disclosure, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to the bus 604. The system 600 may further include one or more of the following components connected to the input / output (I / O) interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output (I / O) interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0131] According to an embodiment of the present disclosure, the method flow according to an embodiment of the present disclosure can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, the above-mentioned functions defined in the system of the embodiment of the present disclosure are executed. According to an embodiment of the present disclosure, the system, equipment, device, module, unit, etc. described above can be implemented by a computer program module.

[0132] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0133] According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium. For example, it may include, but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0134] For example, according to an embodiment of the present disclosure, the computer-readable storage medium may include the ROM 602 and / or the RAM 603 described above and / or one or more memories other than the ROM 602 and the RAM 603 .

[0135] The embodiments of the present disclosure also include a computer program product, which includes a computer program, and the computer program contains program code for executing the method provided by the embodiments of the present disclosure. When the computer program product runs on an electronic device, the program code is used to enable the electronic device to implement the method provided by the embodiments of the present disclosure.

[0136] When the computer program is executed by the processor 601, the above functions defined in the system / device of the embodiment of the present disclosure are executed. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0137] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 609, and / or installed from a removable medium 611. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0138] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0139] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions. It can be understood by those skilled in the art that the features recorded in the various embodiments and / or claims of the present disclosure can be combined and / or combined in a variety of ways, even if such a combination or combination is not explicitly recorded in the present disclosure. In particular, without departing from the spirit and teaching of the present disclosure, the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0140] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. The scope of the present disclosure is defined by the attached claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A license plate recognition method, comprising: Determine the number of license plate characters and license plate character position information of each license plate character according to the license plate type of the license plate in the image; Performing character recognition on the license plate area in the image to obtain a predicted character sequence and predicted character position information of each predicted character in the predicted character sequence; In response to determining that the number of predicted characters of the predicted character sequence is less than the number of characters of the license plate, determining missing character position information; Performing character recognition on the missing character region represented by the missing character position information to obtain a missing character recognition result; as well as The license plate character recognition result of the license plate is determined according to the predicted character sequence, the missing character recognition result and the missing character position information.

2. The method according to claim 1, wherein: Determining the missing character position information includes: The missing character position information is determined according to the character information of the predicted character sequence.

3. The method according to claim 2, wherein: Determining the missing character position information according to the character information of the predicted character sequence includes: In response to determining that the predicted character sequence is missing a province character representing a province, the missing character position information is determined based on the province character area position information representing the province character in the license plate character position information.

4. The method according to claim 2, wherein: Determining the missing character position information according to the character information of the predicted character sequence includes: In response to determining that the predicted character sequence is missing English letters, the missing character position information is determined based on English letter area position information representing the English letters in the license plate character position information.

5. The method according to any one of claims 1 to 4, wherein: Determining the missing character position information includes: The missing character position information is determined according to at least one of the license plate character position information, the predicted character position information and the license plate type.

6. The method according to claim 5, wherein: The predicted character sequence includes N predicted characters, where N is a positive integer; and determining the missing character position information according to at least one of the license plate character position information, the predicted character position information and the license plate type includes: Determining an average license plate character width of the license plate characters according to the number of license plate characters and the license plate character position information; Determining a license plate character spacing threshold between adjacent license plate characters according to the average license plate character width and the license plate type; In response to determining that the distance between the nth predicted character and the n+1th predicted character is greater than the license plate character spacing threshold, obtaining nth predicted character position information of the nth predicted character and n+1th predicted character position information of the n+1th predicted character, where n is an integer greater than or equal to 1 and less than N; and The first target license plate character position information located between the nth predicted character position information and the (n+1)th predicted character position information is determined as the missing character position information.

7. The method according to claim 6, wherein: Determining the license plate character interval threshold between adjacent license plate characters according to the average license plate character width and the license plate type comprises: In response to determining that the license plate type is a single-row license plate, determining a license plate character interval threshold between first adjacent license plate characters as a first interval threshold, wherein the first adjacent license plate characters are the second license plate character and the third license plate character; and Determine a license plate character interval threshold between second adjacent license plate characters as a second interval threshold, wherein the second adjacent license plate characters are other adjacent license plate characters in the license plate characters of the single-row license plate except the first adjacent license plate characters, and the first interval threshold is greater than the second interval threshold.

8. The method according to claim 5, wherein: Determining the missing character position information according to at least one of the license plate character position information, the predicted character position information and the license plate type comprises: In response to determining that the license plate type is a two-row license plate, determining, based on the predicted character position information, center position information of each predicted character in the first row of predicted characters, and at least one of left boundary position information and right boundary position information of each predicted character in the second row of predicted characters; and The missing character position information is determined according to at least one of the left boundary position information and the right boundary position information and the center position information.

9. The method according to claim 5, wherein: Determining the missing character position information according to at least one of the license plate character position information, the predicted character position information and the license plate type comprises: According to the predicted character position information, obtaining the Nth predicted character position information of the Nth predicted character; According to the license plate character position information, obtaining second target license plate character position information of the last license plate character; and In response to determining, based on the Nth predicted character position information and the second target license plate character position information, that the overlap between the Nth predicted character and the last license plate character is less than a preset threshold, the second target license plate character position information is determined as the missing character position information.

10. The method according to claim 1, wherein: Determining the license plate character recognition result of the license plate according to the predicted character sequence, the missing character recognition result and the missing character position information comprises: According to the missing character position information, the missing character sequence is inserted into the corresponding position of the predicted character sequence to obtain the license plate character recognition result.

11. A license plate recognition device, comprising: A license plate character information determination module, used to determine the number of license plate characters of the license plate and the license plate character position information of each license plate character according to the license plate type of the license plate in the image; A predicted character information determination module, used to perform character recognition on the license plate area in the image to obtain a predicted character sequence and predicted character position information of each predicted character in the predicted character sequence; a missing character position information determination module, configured to determine missing character position information in response to determining that the number of predicted characters of the predicted character sequence is less than the number of characters of the license plate; A missing character recognition module is used to perform character recognition on the missing character area represented by the missing character position information to obtain a missing character recognition result; as well as The license plate character recognition result determination module is used to determine the license plate character recognition result of the license plate according to the predicted character sequence, the missing character recognition result and the missing character position information.

12. An electronic device comprising: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 10.

13. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enables the processor to implement the method according to any one of claims 1 to 10.

14. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 10.

Images