A license plate recognition method and related device

Abstract

This application relates to the field of image recognition technology. The embodiments of this application provide a license plate recognition method and related apparatus. After front-end video stream recognition, the method determines different processing methods by judging whether the license plate number recognized by the front-end video stream is in the questionable license plate database, the trusted license plate database, and the error correction license plate database. It can directly identify and determine the license plate number based on the trusted license plate database and correct the license plate number based on the error correction license plate database, thereby improving the accuracy of license plate recognition.

Description

Technical Field

[0001] This application belongs to the field of image recognition technology, and in particular relates to a license plate recognition method and related apparatus. Background Technology

[0002] With the development of modern technology, license plate recognition algorithms are increasingly being used in parking lots to identify vehicles without human intervention, making it a growing trend. Currently, license plate recognition algorithms utilize front-end video streams, with each exit / entrance operating independently. The main factors affecting the accuracy of license plate recognition include the algorithm, camera performance, environment, and weather conditions; different systems also have varying accuracy rates. Therefore, the accuracy of license plate recognition is crucial for the stable operation of unattended parking lots.

[0003] However, current license plate recognition technology is limited to front-end video stream recognition. When image / video recognition technology reaches a bottleneck, it is impossible to further improve the accuracy of license plate recognition algorithms. Summary of the Invention

[0004] This application provides a license plate recognition method and related apparatus. After front-end video stream recognition, it further determines different processing methods by judging whether the license plate number recognized by the front-end video stream is in the questionable license plate database, the trusted license plate database, and the error correction license plate database. It can directly identify and determine the license plate number based on the trusted license plate database and correct the license plate number based on the error correction license plate database, thereby improving the accuracy of license plate recognition.

[0005] In a first aspect, embodiments of this application provide a license plate recognition method, including:

[0006] Obtain the license plate recognition result to be processed, which includes the first license plate number and the license plate image;

[0007] If the first license plate number is in the database of questionable license plates, then determine whether the first license plate number is correct;

[0008] If the first license plate number is correct, then the first license plate number is determined as the license plate recognition result;

[0009] If the first license plate number is incorrect, the second license plate number corresponding to the license plate image is re-determined and the second license plate number is determined as the license plate recognition result;

[0010] If the first license plate number is in the trusted license plate database, then the first license plate number is determined to be the license plate recognition result;

[0011] If the first license plate number is in the error correction license plate database, then the error correction license plate number corresponding to the first license plate number in the error correction license plate database is determined as the license plate recognition result;

[0012] Among them, the questionable license plate database, the trusted license plate database, and the error-correcting license plate database are all databases containing at least one license plate number.

[0013] In conjunction with the first aspect, in one implementation of the embodiments of this application, the method further includes:

[0014] If the first license plate number is correct, increase the trust score corresponding to the first license plate number;

[0015] When the trust score corresponding to the first license plate number is greater than the first preset threshold, the first license plate number is added to the trusted license plate database.

[0016] In conjunction with the first aspect, in one implementation of the embodiments of this application, the method further includes:

[0017] If the first license plate number is incorrect, then after re-determining the second license plate number corresponding to the license plate image, add the error correction score corresponding to the association between the first license plate number and the second license plate number;

[0018] When the error correction score is greater than the second preset threshold, the association between the first license plate number and the second license plate number is added to the error correction license plate database.

[0019] In conjunction with the first aspect, in one implementation of the embodiments of this application, the method further includes:

[0020] If the first license plate number is not in the questioned license plate database, trusted license plate database, or error-corrected license plate database, the third license plate number of the license plate image will be determined through a cloud-based recognition algorithm.

[0021] If the third license plate number is the same as the first license plate number, the trust score corresponding to the first license plate number is increased, and the first license plate number is determined as the license plate recognition result;

[0022] When the trust score corresponding to the first license plate number is greater than the first preset threshold, the first license plate number is added to the trusted license plate database.

[0023] In conjunction with the first aspect, in one implementation of this application embodiment, when determining the third license plate number using a cloud-based recognition algorithm, the confidence level corresponding to the third license plate number is also determined. The method further includes:

[0024] If the third license plate number is different from the first license plate number, and the confidence level is greater than or equal to the preset confidence threshold, then the trust score corresponding to the third license plate number is increased.

[0025] When the trust score corresponding to the third license plate number is greater than the first preset threshold, the third license plate number will be added to the trusted license plate database.

[0026] In conjunction with the first aspect, in one implementation of the embodiments of this application, the method further includes:

[0027] If the third license plate number is different from the first license plate number and the confidence level is less than the preset confidence level threshold, then determine whether the third license plate number is in the trusted license plate database or the error correction license plate database.

[0028] If the third license plate number is in the trusted license plate database, then the third license plate number is determined to be the license plate recognition result, and the error correction score corresponding to the association between the first license plate number and the third license plate number is added;

[0029] When the error correction score corresponding to the association between the first license plate number and the third license plate number is greater than the second preset threshold, the association between the first license plate number and the third license plate number is added to the error correction license plate database.

[0030] If the third license plate number is in the error correction license plate database, the license plate recognition result is determined based on the correlation in the error correction license plate database.

[0031] In conjunction with the first aspect, in one implementation of the embodiments of this application, the method further includes:

[0032] If the third license plate number is not in the trusted license plate database and is not in the error-corrected license plate database, then determine whether the first license plate number is correct;

[0033] If the first license plate number is correct, then the first license plate number is determined as the license plate recognition result;

[0034] If the first license plate number is incorrect, the second license plate number corresponding to the license plate image is re-determined and identified as the license plate recognition result.

[0035] Secondly, embodiments of this application provide a license plate recognition device, characterized in that it includes:

[0036] The acquisition module is used to acquire the license plate recognition result to be processed, which includes the first license plate number and the license plate image;

[0037] The processing module is used to determine whether the first license plate number is correct if it is in the suspected license plate database.

[0038] If the first license plate number is correct, then the first license plate number is determined as the license plate recognition result;

[0039] If the first license plate number is incorrect, the second license plate number corresponding to the license plate image is re-determined and the second license plate number is determined as the license plate recognition result;

[0040] If the first license plate number is in the trusted license plate database, then the first license plate number is determined to be the license plate recognition result;

[0041] If the first license plate number is in the error correction license plate database, then the error correction license plate number corresponding to the first license plate number in the error correction license plate database is determined as the license plate recognition result;

[0042] Among them, the questionable license plate database, the trusted license plate database, and the error-correcting license plate database are all databases containing at least one license plate number.

[0043] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method as described in the first aspect.

[0044] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the method as described in the first aspect.

[0045] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the license plate recognition method described in any of the first aspects above.

[0046] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0047] The beneficial effects of the embodiments in this application compared with the prior art are:

[0048] In this embodiment, after front-end video stream recognition, different processing methods are determined by judging whether the license plate number recognized by the front-end video stream is in the questionable license plate database, the trusted license plate database, and the error correction license plate database. The license plate number can be directly identified and determined based on the trusted license plate database, and the license plate number can be corrected based on the error correction license plate database, thereby improving the accuracy of license plate recognition. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a schematic diagram illustrating an application scenario of an embodiment of this application;

[0051] Figure 2 A flowchart illustrating a license plate recognition method provided in this application embodiment;

[0052] Figure 3 This is an interface diagram of the terminal device 103 displaying a license plate image in an embodiment of this application.

[0053] Figure 4Signaling diagrams provided for embodiments of this application;

[0054] Figure 5 This is a schematic diagram of a license plate recognition device according to an embodiment of this application;

[0055] Figure 6 This is a schematic diagram of a license plate recognition device provided in an embodiment of this application. Detailed Implementation

[0056] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0057] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0058] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0059] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0060] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0061] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0062] The license plate recognition method provided in this application embodiment can be applied to a cloud-based recognition server. This cloud-based recognition server can be used for secondary recognition of the front-end recognition results. Therefore, this cloud-based recognition server is generally connected to the front-end recognition server and has a cloud-based agent recognition service. Specifically, Figure 1 This is a schematic diagram illustrating an application scenario of an embodiment of this application.

[0063] like Figure 1 As shown, the front-end recognition server 101 is connected to the cloud recognition server 102, and the cloud recognition server 102 can be connected to the terminal device 103. Wherein:

[0064] The front-end recognition server 101 is used for front-end recognition of license plate numbers. It is understood that cameras are typically installed at the entrances and exits of unattended parking lots to capture video streams. After the video stream is transmitted to the front-end recognition server 101, the server can analyze it to obtain the corresponding license plate number. In this embodiment, the video stream can be considered as composed of multiple frames of images; therefore, the front-end recognition server 101 can also be considered as processing multiple frames of image data. Whether processing a video stream, a single video clip, or multiple frames of image data, all fall within the protection scope of this embodiment. The front-end recognition server 101 can send the license plate image data and the recognized license plate number to the cloud recognition server 102.

[0065] The cloud-based recognition server 102 is used to perform secondary recognition on the license plate image data and the recognized license plate number, that is, to execute the license plate recognition method provided in this application embodiment. In addition, the cloud-based recognition server 102 can also interact with the terminal device 103 to realize cloud-based agent recognition.

[0066] Terminal device 103 is used to implement cloud-based agent identification. When cloud-based agent identification is required, the cloud-based identification server 102 can be used through terminal device 103 and staff. The specific process is as follows: The cloud-based identification server 102 can send the license plate image and license plate number to terminal device 103. Then, staff can browse the license plate image on terminal device 103 and determine whether the license plate number corresponding to the image is correct. If correct, confirmation information is returned through terminal device 103; if incorrect, error feedback is sent through terminal device 103 and the correct license plate number is entered.

[0067] Based on the above application scenarios, this application provides a license plate recognition method that can be executed by the cloud recognition server 102. Figure 2 A flowchart of a license plate recognition method provided in this application embodiment includes the following steps:

[0068] 201. Obtain the license plate recognition results to be processed;

[0069] In this embodiment, the license plate recognition result to be processed includes a first license plate number and a license plate image. The license plate recognition result to be processed can be the result of preliminary recognition by the front-end recognition server 101 through video surveillance. This result is generally the recognition of the license plate image to obtain the first license plate number. Then, the front-end recognition server 101 can send the recognition result to the cloud recognition server 102. In some embodiments, the front-end recognition server 101 also calculates the confidence level of the recognition result. Only when the confidence level of the recognition result is lower than a preset threshold is the recognition result sent to the cloud recognition server 102 for further recognition (secondary recognition).

[0070] 202. If the first license plate number is in the database of suspected license plates, then determine whether the first license plate number is correct;

[0071] In this embodiment, the cloud-based identification server 102 can determine whether the first license plate number is in the suspected license plate database based on the obtained first license plate number. The suspected license plate database is a database containing at least one license plate number. License plates in this database are generally highly similar to those in the same parking lot and are easily misidentified. Such license plates are manually identified each time they enter the parking lot. Therefore, if the first license plate number is in the suspected license plate database, it indicates that the first license plate number is easily misidentified, and thus a cloud-based identification service is needed for identification.

[0072] Specifically, when the cloud recognition server 102 detects that the first license plate number is in the suspected license plate database, it can call the cloud seat recognition service to identify and determine whether the first license plate number is correct. During the cloud seat recognition service process, the cloud recognition server 102 can send the license plate image and the first license plate number to the terminal device 103 for display. The staff observes whether the license plate number in the license plate image is the first license plate number, that is, to determine whether the first license plate number is correct. This process specifically includes steps 203 and 204.

[0073] It is understandable that the database of questionable license plates can be created manually or by adding license plate numbers to the database after they meet certain conditions. This application does not limit the specific method used in this embodiment.

[0074] 203. If the first license plate number is correct, then the first license plate number is determined as the license plate recognition result;

[0075] Figure 3 This is an interface diagram of the terminal device 103 displaying a license plate image in an embodiment of this application. Figure 3 As shown, the staff observes the license plate image displayed on the screen of terminal device 103, and then determines whether the first license plate number is correct. If correct, they click the "Correct" button, triggering terminal device 103 to return confirmation information to cloud recognition server 102, enabling cloud recognition server 102 to confirm that the first license plate number is correct. After receiving the confirmation information and confirming that the first license plate number is correct, cloud recognition server 102 can confirm that the first license plate number is the license plate recognition result.

[0076] In practical applications, after the cloud recognition server 102 determines that the first license plate number is the license plate recognition result, it can send the result back to the front-end recognition server 101, so that the front-end recognition server 101 can determine the license plate recognition result and then perform the corresponding deduction service, registration service, etc.

[0077] 204. If the first license plate number is incorrect, then re-determine the second license plate number corresponding to the license plate image and determine the second license plate number as the license plate recognition result;

[0078] like Figure 3 As shown, the staff observes the license plate image displayed on the screen of terminal device 103, then determines whether the first license plate number is correct. If the first license plate number is incorrect, they click the "Error" button and manually enter the correct license plate number (which can be referred to as the second license plate number). Terminal device 103 can feed back the information of the entered second license plate number to cloud recognition server 102, so that cloud recognition server 102 can determine that the first license plate number is incorrect and re-use the second license plate number fed back by terminal device 103 as the license plate recognition result.

[0079] In practical applications, after the cloud recognition server 102 determines that the second license plate number is the license plate recognition result, it can feed back to the front-end recognition server 101, so that the front-end recognition server 101 can determine the license plate recognition result and then perform the corresponding deduction service, registration service, etc.

[0080] In this embodiment of the application, if the first license plate number is not in the suspected license plate database, the cloud recognition server 102 further determines whether the first license plate number is in the trusted license plate database, i.e., step 205.

[0081] 205. If the first license plate number is in the trusted license plate database, then the first license plate number is determined to be the license plate recognition result;

[0082] In this embodiment, the cloud-based recognition server 102 can determine whether the first license plate number is in the trusted license plate database based on the obtained first license plate number. The trusted license plate database is a database containing at least one license plate number, and the license plate numbers in this database are generally considered to be trusted. Therefore, if a license plate number is in the trusted license plate database, it is generally considered to be correct; it's just that the front-end recognition cannot accurately identify it.

[0083] When the cloud recognition server 102 detects that the first license plate number is in the trusted license plate database, it can determine that the first license plate number is the license plate recognition result and can feed it back to the front-end recognition server 101, so that the front-end recognition server 101 can determine the license plate recognition result and then perform operations such as deducting fees and registering services corresponding to the license plate number.

[0084] In this embodiment of the application, if the first license plate number is not in the trusted license plate database, the cloud recognition server 102 further determines whether the first license plate number is in the error correction license plate database, i.e., step 206.

[0085] 206. If the first license plate number is in the error correction license plate database, then the error correction license plate number corresponding to the first license plate number in the error correction license plate database is determined as the license plate recognition result;

[0086] Table 1 shows an example of a license plate error correction database. The database includes a relationship between "original license plate number" and "corrected license plate number." The "original license plate number" represents the license plate number before error correction, and the "corrected license plate number" represents the corrected license plate number. For example, the first column in Table 1 indicates that the correct license plate number for the error message "**: *****6" should be "**: *****7".

[0087] Original license plate number Correcting license plate number **：*****6 **：*****7 **：****3* **：****2* **：*****1 **：*****7 ... ...

[0088] In this embodiment, the cloud-based recognition server 102 can determine whether the first license plate number is in the "original license plate number" of the error-correcting license plate database. If so, it means that the first license plate number can be corrected through the license plate numbers in the error-correcting license plate database, and the cloud-based recognition server 102 can determine that the corrected license plate number corresponding to the first license plate number in the error-correcting license plate database is the license plate recognition result. The cloud-based recognition server 102 can feed back the corresponding corrected license plate number to the front-end recognition server 101, so that the front-end recognition server 101 can determine the license plate recognition result and then perform the corresponding charging service, registration service, etc.

[0089] In this embodiment of the application, if the first license plate number is not in the questioned license plate database, the trusted license plate database, and the error-correcting license plate database, the cloud recognition server 102 determines the third license plate number of the license plate image through the cloud recognition algorithm, i.e., step 207.

[0090] 207. Determine the third license plate number from the license plate image using a cloud-based recognition algorithm;

[0091] In this embodiment, the cloud-based recognition algorithm is an algorithm built into the cloud-based recognition server 102, used to recognize license plate numbers based on license plate images. Specifically, the cloud-based recognition algorithm can be a neural network algorithm or a more complex license plate recognition algorithm with higher recognition accuracy; this embodiment does not limit the specific algorithm used.

[0092] After the cloud recognition server 102 determines the third license plate number of the license plate image through the cloud recognition algorithm, it can compare the third license plate number with the first license plate number, thereby executing step 208 or step 209.

[0093] 208. If the third license plate number is the same as the first license plate number, then the first license plate number is determined as the license plate recognition result;

[0094] In this embodiment, the third license plate number is the same as the first license plate number, indicating that the license plate number is consistent after two recognition checks, so there is a high probability that this license plate number is correct. Therefore, after detecting that the third license plate number is the same as the first license plate number, the cloud recognition server 102 can determine that the first license plate number is the license plate recognition result.

[0095] Furthermore, after the cloud-based recognition server 102 detects that the third license plate number is the same as the first license plate number, it can...

[0096] 209. If the third license plate number is different from the first license plate number, then further determine whether the confidence level corresponding to the third license plate number is greater than or equal to the preset confidence threshold.

[0097] In this embodiment, the confidence level corresponding to the third license plate number can be determined simultaneously by the cloud recognition server 102 when determining the third license plate number of the license plate image using a cloud recognition algorithm. The confidence level corresponding to the third license plate number can represent the probability that the third license plate number is correct.

[0098] 210. If the confidence level corresponding to the third license plate number is greater than or equal to the preset confidence threshold, then the third license plate number is determined as the license plate recognition result.

[0099] In this embodiment of the application, if the confidence level corresponding to the third license plate number is greater than or equal to the preset confidence threshold, it indicates that the probability of the third license plate number being correctly identified is relatively high, which meets the requirements. At this time, the cloud recognition server 102 can determine that the third license plate number is the license plate recognition result.

[0100] 211. If the confidence level corresponding to the third license plate number is less than the preset confidence threshold, then further determine whether the third license plate number is in the trusted license plate database or the error correction license plate database.

[0101] If the confidence level corresponding to the third license plate number is less than the preset confidence threshold, it indicates that the secondary cloud recognition cannot accurately determine the license plate number, and verification is required using a trusted license plate database or an error-correcting license plate database. If the third license plate number is in the trusted license plate database or the error-correcting license plate database, proceed to step 212; if the third license plate number is not in the trusted license plate database or the error-correcting license plate database, proceed to step 213.

[0102] 212. If the third license plate number is in the trusted license plate database or the error correction license plate database, then the third license plate number is determined as the license plate recognition result, or the license plate recognition result is determined based on the association in the error correction license plate database.

[0103] In this embodiment of the application, if the third license plate number is in the trusted license plate database, the cloud recognition server 102 can determine that the third license plate number is the license plate recognition result and can feed it back to the front-end recognition server 101, so that the front-end recognition server 101 can determine the license plate recognition result and then perform the corresponding deduction service, registration service and other operations.

[0104] In this embodiment, if the third license plate number is in the error-correcting license plate database, the cloud-based recognition server 102 can determine the license plate recognition result based on the association in the error-correcting license plate database. For example, if the "original license plate number" is the first license plate number and the corresponding "error-correcting license plate number" is the third license plate number, then the third license plate number is determined to be the license plate recognition result. As another example, if the "original license plate number" is the third license plate number and the corresponding "error-correcting license plate number" is the fourth license plate number, then the fourth license plate number can be determined to be the license plate recognition result.

[0105] 213. If the third license plate number is not in the trusted license plate database or the error correction license plate database, the license plate image will be processed through the cloud seat recognition service.

[0106] In this embodiment, the method by which the cloud recognition server 102 processes license plate images through the cloud seat recognition service is the same as described above. Figure 3 The corresponding embodiments are similar and will not be described in detail here. It is understood that in step 213, the third license plate number can also be determined by cloud-based customer service to verify its correctness. The cloud-based customer service can determine whether the first and third license plate numbers are correct, or the correct license plate number can be manually entered by staff.

[0107] The following provides a detailed description of the updates to the questionable license plate database, the trusted license plate database, and the error-correcting license plate database. License plate numbers can be manually added to these databases, or they can be added by assigning scores to license plate numbers. When a license plate number's score reaches a preset threshold, that license plate number is added to the database. Specific rules include the following:

[0108] In step 203, after the cloud recognition server 102 determines that the first license plate number is the license plate recognition result, the trust score corresponding to the first license plate number can be increased. For example, it can be set that 1 trust score is added each time the first license plate number is identified as the license plate recognition result.

[0109] In step 204, after the cloud recognition server 102 determines that the second license plate number is the license plate recognition result, it can increase the error correction score corresponding to the association between the first license plate number and the second license plate number. For example, if the error correction score of association A (first license plate number and second license plate number) is 0, after the cloud recognition server 102 determines that the second license plate number is the license plate recognition result, it can increase the error correction score of association A by 1, making it 1 point.

[0110] In step 208, after the cloud recognition server 102 determines that the first license plate number is the license plate recognition result, the trust score corresponding to the first license plate number can be increased. For example, it can be set that 1 trust score is added each time the first license plate number is identified as the license plate recognition result.

[0111] In step 210, after the cloud recognition server 102 determines that the third license plate number is the license plate recognition result, it can increase the error correction score corresponding to the association between the first license plate number and the third license plate number. For example, if the error correction score of association B (first license plate number and third license plate number) is 3, after the cloud recognition server 102 determines that the third license plate number is the license plate recognition result, it can increase the error correction score of association A by 1, making it 4 points.

[0112] In step 212, if the third license plate number is in the trusted license plate database, the cloud recognition server 102 can determine that the third license plate number is the license plate recognition result, and then add the error correction score corresponding to the association between the first license plate number and the third license plate number.

[0113] When the cloud-based recognition server 102 processes license plate images through the cloud-based agent recognition service, it can increase the trust score corresponding to the first license plate number after determining that the first license plate number is the license plate recognition result. When it determines that a license plate number other than the first license plate number is the license plate recognition result, it can increase the error correction score between the first license plate number and the recognition result of that license plate.

[0114] In this embodiment, the trust score or error correction score added for each step can be different and can be set according to actual needs. For example, steps 203 and 204 are identified through cloud agent identification, so the trust score or error correction score can be increased significantly.

[0115] The cloud-based recognition server 102 can check the license plate number for each update of the trust score or error correction score. When the trust score corresponding to the license plate number is greater than a first preset threshold, the license plate number is added to the trusted license plate database. When the error correction score corresponding to the association of the license plate number is greater than a second preset threshold, the association is added to the error correction license plate database. For example, when the trust score corresponding to the first license plate number is greater than the first preset threshold, the first license plate number is added to the trusted license plate database. As another example, when the error correction score corresponding to the association between the first and second license plate numbers is greater than the second preset threshold, the association is added to the error correction license plate database.

[0116] In some embodiments, the cloud recognition server 102 can check the score of each license plate number at regular intervals and update the license plate database according to whether the score reaches a preset threshold. This will not be elaborated further in this embodiment.

[0117] Figure 4 This is a signaling diagram provided for an embodiment of this application. After the front-end recognition server 101 performs preliminary license plate recognition, it can obtain the front-end recognition result. Then, the front-end recognition server 101 can send the license plate image and the front-end recognition result to the cloud recognition server 102. The cloud recognition server 102 may include a license plate database module, a cloud recognition module, and a cloud agent recognition module. The license plate database module is mainly used to determine whether the front-end recognition result is in the questionable license plate database, the trusted license plate database, or the error correction license plate database, and to perform corresponding processing. The cloud recognition module is mainly used to perform cloud recognition and then process the results based on the cloud recognition results. The cloud agent recognition module is mainly used for manual identification to determine whether the front-end recognition result is correct.

[0118] Depending on the processing, the cloud-based recognition server 102 can generate a trust instruction or an error correction instruction to the front-end recognition server 101. The trust instruction is generally used to confirm that the front-end recognition result is correct, and the front-end recognition server 101 can directly use the front-end recognition result as the license plate recognition result. The error correction instruction is generally used to correct errors, including cases where the license plate number is correct; the front-end recognition server 101 can use the correct license plate number in the error correction instruction as the license plate recognition result.

[0119] Understandably, when the cloud-based agent recognition module determines that the front-end recognition result is correct, it will increase the trust score of the front-end recognition result. When the cloud-based agent recognition module determines that the front-end recognition result is incorrect and a license plate number is manually entered, it can increase the error correction score corresponding to the association between the front-end recognition result and the manually entered license plate number. When the cloud-based recognition module determines that the cloud-based recognition result is consistent with the front-end recognition result, it can increase the trust score of the front-end recognition result. When the confidence level of the cloud-based recognition result obtained by the cloud-based recognition module reaches a threshold, it can increase the error correction score corresponding to the association between the front-end recognition result and the cloud-based recognition result.

[0120] When the trust score reaches a preset threshold, the cloud recognition server 102 can add the license plate number corresponding to the trust score to the trusted license plate database. When the error correction score reaches a preset threshold, the cloud recognition server 102 can add the association corresponding to the error correction score to the error correction license plate database.

[0121] The license plate recognition method provided in this application not only determines different processing methods by judging whether the initially identified license plate number is in the questionable license plate database, the trusted license plate database, and the error-correcting license plate database, but also improves the accuracy of license plate recognition by directly identifying the license plate number based on the trusted license plate database and correcting the license plate number based on the error-correcting license plate database. Furthermore, by using algorithms to assist cloud-based agent recognition in pre-judgment, the efficiency of license plate recognition can be improved, the workload of cloud agents can be reduced, and thus the goal of cost control can be achieved.

[0122] Figure 5 This is a schematic diagram of a license plate recognition device according to an embodiment of this application. The license plate recognition device 500 includes:

[0123] Get module 501, used to execute or implement Figure 2 Step 201 in the corresponding embodiments;

[0124] Processing module 502 is used to execute or implement Figure 2 Steps 202 to 213 in the corresponding embodiments.

[0125] Figure 6This is a schematic diagram of a license plate recognition device provided in an embodiment of this application. The device 600 includes a memory 602, a processor 601, and a computer program 603 stored in the memory 602 and executable on the processor 601. When the processor 601 executes the computer program 603, it implements... Figure 2 or Figure 4 The methods of the corresponding embodiments.

[0126] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0127] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0128] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0129] This application provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps described in the above-described method embodiments.

[0130] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographic device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0131] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0132] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0133] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0134] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0135] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A license plate recognition method, characterized in that, include: Obtain the license plate recognition result to be processed, the license plate recognition result to be processed includes a first license plate number and a license plate image; If the first license plate number is in the database of suspected license plates, then determine whether the first license plate number is correct; If the first license plate number is correct, then the first license plate number is determined to be the license plate recognition result; If the first license plate number is incorrect, then the second license plate number corresponding to the license plate image is re-determined and the second license plate number is determined as the license plate recognition result; If the first license plate number is in the trusted license plate database, then the first license plate number is determined to be the license plate recognition result; If the first license plate number is in the error correction license plate database, then the error correction license plate number corresponding to the first license plate number in the error correction license plate database is determined as the license plate recognition result; If the first license plate number is not in the questioned license plate database, the trusted license plate database, and the error-corrected license plate database, then the third license plate number of the license plate image is determined by a cloud-based recognition algorithm. If the third license plate number is the same as the first license plate number, then the trust score corresponding to the first license plate number is increased, and the first license plate number is determined to be the license plate recognition result; When the trust score corresponding to the first license plate number is greater than the first preset threshold, the first license plate number is added to the trusted license plate database. The questioned license plate database, the trusted license plate database, and the error-correcting license plate database are all databases containing at least one license plate number.

2. The method as described in claim 1, characterized in that, If the first license plate number is in the suspected license plate database, after determining whether the first license plate number is correct, the method further includes: If the first license plate number is correct, then increase the trust score corresponding to the first license plate number; When the trust score corresponding to the first license plate number is greater than the first preset threshold, the first license plate number is added to the trusted license plate database.

3. The method as described in claim 1, characterized in that, If the first license plate number is in the suspected license plate database, after determining whether the first license plate number is correct, the method further includes: If the first license plate number is incorrect, then after re-determining the second license plate number corresponding to the license plate image, an error correction score corresponding to the association between the first license plate number and the second license plate number is added; When the error correction score is greater than the second preset threshold, the association between the first license plate number and the second license plate number is added to the error correction license plate database.

4. The method as described in claim 1, characterized in that, When determining the third license plate number using a cloud-based recognition algorithm, the method also determines the confidence level corresponding to the third license plate number. The method further includes: If the third license plate number is different from the first license plate number, and the confidence level is greater than or equal to a preset confidence threshold, then the trust score corresponding to the third license plate number is increased. When the trust score corresponding to the third license plate number is greater than the first preset threshold, the third license plate number is added to the trusted license plate database.

5. The method as described in claim 4, characterized in that, The method further includes: If the third license plate number is different from the first license plate number, and the confidence level is less than the preset confidence threshold, then it is determined whether the third license plate number is in the trusted license plate database or the error correction license plate database. If the third license plate number is in the trusted license plate database, then the third license plate number is determined to be a license plate recognition result, and the error correction score corresponding to the association between the first license plate number and the third license plate number is increased. When the error correction score corresponding to the association between the first license plate number and the third license plate number is greater than the second preset threshold, the association between the first license plate number and the third license plate number is added to the error correction license plate database. If the third license plate number is in the error correction license plate database, the license plate recognition result is determined based on the association in the error correction license plate database.

6. The method as described in claim 5, characterized in that, The method further includes: If the third license plate number is not in the trusted license plate database and not in the error correction license plate database, then it is determined whether the first license plate number is correct. If the first license plate number is correct, then the first license plate number is determined to be the license plate recognition result; If the first license plate number is incorrect, then the second license plate number corresponding to the license plate image is re-determined and the second license plate number is determined as the license plate recognition result.

7. A license plate recognition device, characterized in that, include: The acquisition module is used to acquire the license plate recognition result to be processed, which includes a first license plate number and a license plate image; The processing module is used to perform the following operations: If the first license plate number is in the database of suspected license plates, then determine whether the first license plate number is correct; If the first license plate number is correct, then the first license plate number is determined to be the license plate recognition result; If the first license plate number is incorrect, then the second license plate number corresponding to the license plate image is re-determined and the second license plate number is determined as the license plate recognition result; If the first license plate number is in the trusted license plate database, then the first license plate number is determined to be the license plate recognition result; If the first license plate number is in the error correction license plate database, then the error correction license plate number corresponding to the first license plate number in the error correction license plate database is determined as the license plate recognition result; If the first license plate number is not in the questioned license plate database, the trusted license plate database, and the error-corrected license plate database, then the third license plate number of the license plate image is determined by a cloud-based recognition algorithm. If the third license plate number is the same as the first license plate number, then the trust score corresponding to the first license plate number is increased, and the first license plate number is determined to be the license plate recognition result; When the trust score corresponding to the first license plate number is greater than the first preset threshold, the first license plate number is added to the trusted license plate database. The questioned license plate database, the trusted license plate database, and the error-correcting license plate database are all databases containing at least one license plate number.

8. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 6.

Images