Driver's license information recognition method, device, electronic device and storage medium

Abstract

The present invention relates to the field of image processing technology, and provides a method, device, electronic device and storage medium for identifying driver's license information. The driver's license information identification method comprises: inputting a driver's license picture to be identified into a key point detection network to obtain the contour of the driver's license picture and the position of a feature area; adjusting the driver's license picture according to the contour and the position of the feature area; inputting the adjusted driver's license picture into a text box detection network to obtain the text box position of the adjusted driver's license picture; inputting the adjusted driver's license picture and the text box position into a text recognition network to obtain the text information of the adjusted driver's license picture; and verifying the text information and outputting the text information with correct verification. The present invention can quickly realize the authenticity identification of the driver's license, and identify and verify all the information in the driver's license, and finally output the driver's license information with correct verification.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a driver's license information recognition method, device, electronic equipment and storage medium. Background Art

[0002] With the development of the OTA (Online Travel Agent) industry, the car rental business continues to expand and the number of driver's license reviews is increasing.

[0003] Some of the existing driver's license recognition algorithms lack verification of the authenticity of the driver's license information, and cannot implement driver's license inspection well. Instead, they increase the difficulty of business review and reduce user experience. Some can only recognize basic information such as identity information, and cannot recognize all the information of the driver's license. They cannot be applied to all scenarios and have poor algorithm reusability.

[0004] It can be seen that how to quickly, accurately and comprehensively review driver's licenses and reduce manual operating costs has become an urgent problem that needs to be solved.

[0005] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0006] In view of this, the present invention provides a driver's license information recognition method, device, electronic device and storage medium, which can quickly realize the authenticity recognition of the driver's license, and identify and verify all the information in the driver's license, and finally output the verified correct driver's license information.

[0007] One aspect of the present invention provides a method for identifying driver's license information, comprising: inputting a driver's license picture to be identified into a key point detection network to obtain the contour and feature area position of the driver's license picture; adjusting the driver's license picture according to the contour and the feature area position; inputting the adjusted driver's license picture into a text box detection network to obtain the text box position of the adjusted driver's license picture; inputting the adjusted driver's license picture and the text box position into a text recognition network to obtain text information of the adjusted driver's license picture; and verifying the text information and outputting the verified text information.

[0008] In some embodiments, the feature area corresponds to a seal area in the driver's license image.

[0009] In some embodiments, obtaining the contour and position of the feature area of ​​the driver's license image includes: extracting the contour and position of the feature area of ​​the driver's license image through the key point detection network to obtain multiple sets of coordinate information; judging whether the feature area is complete based on the coordinate information of the feature area and a preset missing threshold; if so, outputting the contour and the position of the feature area; if not, outputting the recognition result of the abnormal driver's license.

[0010] In some embodiments, adjusting the driver's license picture includes: performing angle correction, affine transformation and cropping on the driver's license picture, and outputting the adjusted driver's license picture.

[0011] In some embodiments, the text box detection network includes a text detection algorithm and a post-processing merging algorithm, the post-processing merging algorithm is a local tilt-aware non-maximum suppression algorithm, and obtaining the text box position of the adjusted driver's license image includes: obtaining all text boxes of the adjusted driver's license image through the text detection algorithm; merging the text boxes through the post-processing merging algorithm, and outputting the merged text box position.

[0012] In some embodiments, the text recognition network includes a text recognition algorithm and a post-processing error correction algorithm, and obtaining the text information of the adjusted driver's license picture includes: obtaining all text information of the adjusted driver's license picture through the text recognition algorithm; correcting the text information through the post-processing error correction algorithm, and outputting the corrected text information.

[0013] In some embodiments, the verifying of the text information further includes: if the verification is wrong, outputting a recognition result of an abnormal driver's license.

[0014] Another aspect of the present invention provides a driver's license information recognition device, which is used to implement the driver's license information recognition method described in any of the above embodiments, and the driver's license information recognition device includes: a key point detection module, configured to input the driver's license image to be identified into a key point detection network, and obtain the contour and feature area position of the driver's license image; a picture adjustment module, configured to adjust the driver's license image according to the contour and the position of the feature area; a text box detection module, configured to input the adjusted driver's license image into a text box detection network, and obtain the text box position of the adjusted driver's license image; a text recognition module, configured to input the adjusted driver's license image and the text box position into a text recognition network, and obtain the text information of the adjusted driver's license image; and a text verification module, configured to verify the text information and output the verified text information.

[0015] Another aspect of the present invention provides an electronic device, comprising: a processor; a memory, wherein the memory stores executable instructions; wherein when the executable instructions are executed by the processor, the driver's license information recognition method described in any of the above embodiments is implemented.

[0016] Another aspect of the present invention provides a computer-readable storage medium for storing a program, which, when executed, implements the driver's license information recognition method described in any of the above embodiments.

[0017] The beneficial effects of the present invention compared with the prior art include at least:

[0018] Through the key point detection network, the contour and feature area position of the driver's license image are obtained, and then the authenticity of the driver's license image can be quickly identified through the feature area;

[0019] By adjusting the driver's license image and inputting it into the text box detection network, the accurate text box position is obtained, and then according to the adjusted driver's license image and text box position, all the text information on the driver's license image is obtained by using the text recognition network, and then verified, and finally the verified correct driver's license information is output;

[0020] Therefore, the driver's license information recognition method, device, electronic device and storage medium of the present invention can greatly improve the insufficient capabilities of existing recognition algorithms for the recognition scenarios of massive driver's license information in the OTA industry, and achieve high-accuracy, rapid batch driver's license authenticity verification and information recognition, greatly improve the recognition efficiency of driver's license information, reduce labor costs, and improve review production efficiency.

[0021] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present invention, and together with the specification are used to explain the principles of the present invention. Obviously, the accompanying drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0023] Figure 1 A diagram showing the main steps of the method for identifying driver's license information in an embodiment of the present invention;

[0024] Figure 2 A specific flow chart showing the method for identifying driver's license information in an embodiment of the present invention is shown;

[0025] Figure 3A schematic diagram of a module of a driver's license information recognition device according to an embodiment of the present invention is shown;

[0026] Figure 4 A schematic diagram showing the structure of an electronic device in an embodiment of the present invention; and

[0027] Figure 5 A schematic diagram showing the structure of a computer-readable storage medium in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0028] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the embodiments set forth herein. Instead, these embodiments are provided to make the present invention comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art.

[0029] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0030] The step numbers in the following embodiments are only used to indicate different execution contents, and do not strictly limit the execution order between the steps. It should be noted that the features in the embodiments of the present invention and in different embodiments can be combined with each other without conflict.

[0031] Figure 1 The main steps of the driver's license information recognition method in the embodiment are shown, referring to Figure 1 As shown, the driver's license information recognition method in this embodiment includes: in step S110, the driver's license picture to be recognized is input into the key point detection network to obtain the contour and feature area position of the driver's license picture; in step S120, the driver's license picture is adjusted according to the contour and feature area position; in step S130, the adjusted driver's license picture is input into the text box detection network to obtain the text box position of the adjusted driver's license picture; in step S140, the adjusted driver's license picture and text box position are input into the text recognition network to obtain the text information of the adjusted driver's license picture; and in step S150, the text information is verified and the verified correct text information is output.

[0032] Among them, the feature area corresponds to the seal area in the driver's license image, such as the seal / photo area of ​​the issuing authority. The feature area contains key information for identifying the authenticity of the driver's license. The authenticity of the driver's license can be quickly identified through the feature area.

[0033] The driver's license information recognition method of the above-mentioned embodiment obtains the contour and feature area position of the driver's license picture through the key point detection network, and then can quickly identify the authenticity of the driver's license picture through the feature area; further, by adjusting the driver's license picture and inputting it into the text box detection network, an accurate text box position is obtained, and then according to the adjusted driver's license picture and text box position, all text information on the driver's license picture is obtained by using the text recognition network, and verified, and finally the verified correctly driver's license information is output; thus, the above-mentioned driver's license information recognition method can be applied to the recognition scenarios of massive driver's license information in the OTA industry, improve the insufficient capabilities of existing recognition algorithms, and achieve high-accuracy, rapid batch driver's license authenticity verification and information recognition, greatly improve the recognition efficiency of driver's license information, reduce labor costs, and improve review production efficiency.

[0034] Combine the following Figure 2 The example of driver's license information recognition shown in the figure describes in detail each step of the driver's license information recognition method. The driver's license information recognition method can be specifically executed by a driver's license information recognition model based on deep learning. The driver's license information recognition model establishes a new algorithm model to address the situation where the existing recognition algorithm is insufficient, introduces a new algorithm module to verify the authenticity of the driver's license and recognize all the information, and realizes high accuracy and rapid batch driver's license authenticity verification and information recognition.

[0035] In step S110, through S110-2, the driver's license picture to be recognized is input into the driver's license information recognition model. The driver's license picture is, for example, an image I with a size of H×W. input ; Through S110-4, the driver's license picture I input Send to the key point detection network Net j , recognize the driver's license image I input contours and feature areas.

[0036] Key point detection network Net j First extract the driver's license image I input The contour and position of the feature area are obtained to obtain multiple sets of coordinate information, for example, n sets of coordinate information are obtained, each set of coordinate information is marked as [x 1k ,y 1k ,x 2k ,y 2k ]. Then, through S110-6, the key point detection network Net jBased on the coordinate information of the feature area and the preset missing threshold, determine whether the feature area is complete. If so, output the contour and the position of the feature area and enter the subsequent steps. Otherwise, execute S200 and output the recognition result of the abnormal driving license.

[0037] Key point detection network Net j The formula for extracting the position of contours and feature areas and obtaining coordinate information is as follows:

[0038] {[x 1k ,y 1k ,x 2k ,y 2k ],…}=Net j (I input ),k=1,2,…,n;

[0039] Determine whether the feature area is complete. Specifically, determine whether the missing coordinate information of the feature area reaches c. If so, it means that the feature area (key information) is seriously missing and the user needs to upload the driver's license picture again; if it is less than c, you can proceed to the next step of driver's license information recognition.

[0040] In step S120, the driver's license image is cropped, rotated, scaled, etc., the driver's license image is rotated to a suitable angle and unnecessary redundant information is cropped, and the adjusted driver's license image is output. Specifically, according to the obtained n sets of coordinate information [x 1k ,y 1k ,x 2k ,y 2k ], and use the image adjustment algorithm Adjust(·) to adjust the driver's license image I input Perform angle correction, affine transformation and cropping to obtain the adjusted driver's license image I with an output size of h×w and an angle of α r .

[0041] The formula of the image adjustment algorithm Adjust(·) is as follows:

[0042] I r =Adjust(I input ,{[x 1k ,y 1k ,x 2k ,y 2k ],…}),k=1,2,…,n.

[0043] In step S130, the adjusted driver's license image is sent to a text box detection network. The text box detection network is modified based on the EAST (Efficient and Accuracy Scene Text) network. The text box detection network includes a text detection algorithm and a post-processing merging algorithm. It can identify all text boxes in the driver's license image and merge some text boxes to obtain the accurate text box position.

[0044] By combining traditional recognition methods and post-processing methods, an improved text box detection network is formed, which can ensure the accuracy of recognition information and greatly improve the recognition speed of the algorithm, effectively improving the business operation efficiency of driver's license information recognition needs in OTA scenarios.

[0045] The post-processing merging algorithm is specifically a local tilt-aware non-maximum suppression (NMS) algorithm. The local tilt-aware NMS algorithm is a variant of the NMS algorithm and can detect tilted text boxes. The process of obtaining the text box position through the text box detection network specifically includes: obtaining all text boxes of the adjusted driver's license image through the text detection algorithm; merging some text boxes through the post-processing merging algorithm, and outputting the merged text box position.

[0046] Specifically, the adjusted driver's license image I r Send to the text box detection network Net d Get m text boxes Then, the local tilt-aware NMS algorithm is used to merge the m text boxes to obtain Q text box coordinates.

[0047] Among them, the text box detection network Net d The formula for identifying text boxes is:

[0048]

[0049] The formula for merging text boxes using the local tilt-aware NMS algorithm is:

[0050]

[0051] In step S140, the adjusted driver's license image and text box position are sent to the text recognition network. The text recognition network is transformed from the CRNN (abbreviation of convolutional neural network CNN + recurrent neural network RNN) network, including a text recognition algorithm and a post-processing error correction algorithm. It can recognize all the text information in the driver's license image and correct the misrecognized text to obtain all the text information on the driver's license image.

[0052] By combining traditional recognition methods and post-processing methods to form an improved text detection network, the accuracy of the recognition information can be ensured, the recognition speed of the algorithm can be greatly improved, and the business operation efficiency of the driver's license information recognition needs in the OTA scenario can be effectively improved.

[0053] The process of obtaining text information through the text recognition network specifically includes: obtaining all text information of the adjusted driver's license image through the text recognition algorithm; correcting the text information through the post-processing error correction algorithm, and outputting the corrected text information. Among them, error correction includes correcting some repeated recognized texts, correcting obvious recognition errors, etc.

[0054] Specifically, the obtained Q text box coordinates and the adjusted driver's license image I r Send to the text recognition network Net r In the example above, we obtain Q text recognition results:

[0055]

[0056] In step S150, through S150-2, the text information is analyzed and verified, and the accuracy of the text information can be verified in conjunction with the layout analysis. If the verification is correct, the driver's license information of the corresponding field is output. If the verification is wrong, the recognition result of the driver's license abnormality is output, and the user needs to re-upload the driver's license picture. The algorithm of self-verification of the driver's license information can avoid errors in the recognition of the driver's license information and ensure the accuracy of the recognition information.

[0057] Specifically, the Q text recognition results obtained are sent to the verification algorithm Cail(·), and the recognized text information is mutually and comprehensively verified (for example, the number of digits in the ID card number, the gender digit, etc.). If the verification passes, the text information is output. Otherwise, the driver's license is judged to be abnormal.

[0058] The formula of the verification algorithm Cail(·) is:

[0059] bool=Cail({r 1 ,r 2 ,…,r j}),j=1,2,…,Q

[0060]

[0061] In the above embodiments, some algorithm principles that are not specifically described are existing technologies and are therefore not described in detail.

[0062] The above-mentioned driver's license information recognition method, when applied to business scenarios such as car rental in the OTA industry, can effectively solve the problem that the existing algorithm cannot recognize all the information in the driver's license, and cannot realize the authenticity verification and information self-verification of the driver's license. Through the above-mentioned driver's license information recognition method, the authenticity of the driver's license image can be automatically identified, and the required driver's license information can be automatically extracted and verified. The recognition accuracy is high and the speed is fast, which greatly improves the recognition efficiency of the driver's license information; through the driver's license authenticity recognition and all information recognition, it can completely replace the manual review work, greatly save the labor operation cost, increase the review rate, and improve the user experience.

[0063] An embodiment of the present invention further provides a driver's license information recognition device, which can be used to implement the driver's license information recognition method described in any of the above embodiments. Figure 3 The main modules of the driver's license information recognition device in the embodiment are shown, referring to Figure 3 As shown, the driver's license information recognition device 300 in this embodiment includes: a key point detection module 310, configured to input the driver's license image to be recognized into a key point detection network, and obtain the contour and feature area position of the driver's license image; a picture adjustment module 320, configured to adjust the driver's license image according to the contour and feature area position; a text box detection module 330, configured to input the adjusted driver's license image into a text box detection network, and obtain the text box position of the adjusted driver's license image; a text recognition module 340, configured to input the adjusted driver's license image and text box position into a text recognition network, and obtain the text information of the adjusted driver's license image; and a text verification module 350, configured to verify the text information and output the verified correct text information.

[0064] Furthermore, the driver's license information recognition device 300 may also include modules for implementing other process steps of the above-mentioned driver's license information recognition method embodiments. The specific principles of each module can refer to the description of the above-mentioned driver's license information recognition method embodiments, and will not be repeated here.

[0065] As described above, the driver's license information recognition device of the present invention can obtain the outline and feature area position of the driver's license picture through the key point detection network, and then can quickly identify the authenticity of the driver's license picture through the feature area; further, by adjusting the driver's license picture and inputting it into the text box detection network, an accurate text box position is obtained, and then according to the adjusted driver's license picture and text box position, all text information on the driver's license picture is obtained using the text recognition network, and verified, and finally the verified correctly driver's license information is output; thus, the driver's license information recognition device of the present invention can greatly improve the recognition efficiency of the driver's license information, automatically realize the authenticity verification of the driver's license and the recognition of all information, reduce labor costs, and improve the review production efficiency.

[0066] An embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein executable instructions are stored in the memory, and when the executable instructions are executed by the processor, the driver's license information recognition method described in any of the above embodiments is implemented.

[0067] As described above, the electronic device of the present invention can obtain the outline and feature area position of the driver's license picture through the key point detection network, and then can quickly identify the authenticity of the driver's license picture through the feature area; further, by adjusting the driver's license picture and inputting it into the text box detection network, an accurate text box position is obtained, and then according to the adjusted driver's license picture and text box position, all text information on the driver's license picture is obtained using the text recognition network, and verified, and finally the verified correctly driver's license information is output; thus, the electronic device of the present invention can greatly improve the recognition efficiency of the driver's license information, automatically realize the authenticity verification of the driver's license and the recognition of all information, reduce labor costs, and improve the review production efficiency.

[0068] Figure 4 is a schematic diagram of the structure of an electronic device in an embodiment of the present invention. It should be understood that: Figure 4 The modules are only schematically shown, and these modules may be virtual software modules or actual hardware modules. The combination and splitting of these modules and the addition of other modules are all within the protection scope of the present invention.

[0069] like Figure 4 As shown, the electronic device 400 is in the form of a general computing device. The components of the electronic device 400 include but are not limited to: at least one processing unit 410, at least one storage unit 420, a bus 430 connecting different platform components (including the storage unit 420 and the processing unit 410), a display unit 440, etc.

[0070] The storage unit stores program codes, which can be executed by the processing unit 410, so that the processing unit 410 executes the steps of the driver's license information recognition method described in any of the above embodiments. For example, the processing unit 410 can execute the following steps: Figure 1 and Figure 2 Steps shown.

[0071] The storage unit 420 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 4201 and / or a cache storage unit 4202 , and may further include a read-only storage unit (ROM) 4203 .

[0072] The storage unit 420 may also include a program / utility 4204 having one or more program modules 4205, such program modules 4205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0073] Bus 430 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0074] The electronic device 400 can also communicate with one or more external devices 500, which can be one or more of a keyboard, a pointing device, a Bluetooth device, etc. These external devices 500 enable the user to communicate interactively with the electronic device 400. The electronic device 400 can also communicate with one or more other computing devices, and the computer device shown includes a router and a modem. This communication can be performed through an input / output (I / O) interface 450. In addition, the electronic device 400 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN) and / or a public network, such as the Internet) through a network adapter 460. The network adapter 460 can communicate with other modules of the electronic device 400 through a bus 430. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms.

[0075] The embodiment of the present invention further provides a computer-readable storage medium for storing a program, which, when executed, implements the driver's license information identification method described in any of the above embodiments. In some possible implementations, various aspects of the present invention may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the driver's license information identification method described in any of the above embodiments.

[0076] As described above, the computer-readable storage medium of the present invention can obtain the outline and feature area position of the driver's license picture through the key point detection network, and then quickly identify the authenticity of the driver's license picture through the feature area; further, by adjusting the driver's license picture and inputting it into the text box detection network, an accurate text box position is obtained, and then according to the adjusted driver's license picture and text box position, all text information on the driver's license picture is obtained using the text recognition network, and verified, and finally the verified correctly driver's license information is output; thus, the computer-readable storage medium of the present invention can greatly improve the recognition efficiency of the driver's license information, automatically realize the authenticity verification of the driver's license and the recognition of all information, reduce labor costs, and improve the review production efficiency.

[0077] Figure 5 Schematic diagram of the structure of the computer-readable storage medium of the present invention. Figure 5 As shown, a program product 600 for implementing the above method according to an embodiment of the present invention is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0078] The program product may adopt any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media include, but are not limited to: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0079] Computer-readable storage media may include data signals propagated in baseband or as part of a carrier wave, wherein readable program codes are carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0080] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may 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 may be connected to an external computing device, such as through the Internet using an Internet service provider.

[0081] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A method for identifying driver's license information, characterized in that: include: Inputting the driver's license image to be identified into a key point detection network to obtain the contour and the position of the feature area of ​​the driver's license image; Wherein, the characteristic region corresponds to the seal region in the driver's license image, and the obtaining of the contour of the driver's license image and the position of the characteristic region includes: extracting the contour of the driver's license image and the position of the characteristic region through the key point detection network to obtain multiple sets of coordinate information; judging whether the characteristic region is complete according to the coordinate information of the characteristic region and a preset missing threshold; if so, outputting the contour and the position of the characteristic region; if not, outputting the recognition result of the abnormal driver's license; Adjusting the driver's license image according to the contour and the position of the feature area; Inputting the adjusted driver's license picture into a text box detection network to obtain a text box position of the adjusted driver's license picture; Inputting the adjusted driver's license picture and the text box position into a text recognition network to obtain text information of the adjusted driver's license picture; and The text information is verified and the text information with correct verification is output.

2. The method for identifying driver's license information according to claim 1, wherein: The adjusting the driver's license picture includes: The driver's license image is subjected to angle correction, affine transformation and cropping, and the adjusted driver's license image is output.

3. The method for identifying driver's license information according to claim 1, wherein: The text box detection network includes a text detection algorithm and a post-processing merging algorithm, wherein the post-processing merging algorithm is a local tilt-aware non-maximum suppression algorithm. The step of obtaining the text box position of the adjusted driver's license image includes: Obtaining all text boxes of the adjusted driver's license image through the text detection algorithm; The text boxes are merged by the post-processing merging algorithm, and the position of the merged text boxes is output.

4. The method for identifying driver's license information according to claim 1, wherein: The text recognition network includes a text recognition algorithm and a post-processing error correction algorithm, and the obtaining of the text information of the adjusted driver's license image includes: Obtaining all text information of the adjusted driver's license image through the text recognition algorithm; The text information is corrected by the post-processing error correction algorithm, and the corrected text information is output.

5. The method for identifying driver's license information according to claim 1, wherein: The verifying of the text information further includes: If the verification is wrong, the recognition result of abnormal driver's license will be output.

6. A driver's license information recognition device, used to implement the driver's license information recognition method according to any one of claims 1 to 5, characterized in that: The driver's license information recognition device comprises: A key point detection module, configured to input the driver's license image to be identified into a key point detection network, and obtain the contour and the position of the feature area of ​​the driver's license image; The characteristic region corresponds to the seal region in the driver's license image, and the key point detection module obtains the contour of the driver's license image and the position of the characteristic region, including: extracting the contour and the position of the characteristic region of the driver's license image through the key point detection network to obtain multiple sets of coordinate information; judging whether the characteristic region is complete according to the coordinate information of the characteristic region and a preset missing threshold; if so, outputting the contour and the position of the characteristic region; if not, outputting the recognition result of the abnormal driver's license; An image adjustment module, configured to adjust the driver's license image according to the position of the contour and the feature area; A text box detection module, configured to input the adjusted driver's license image into a text box detection network to obtain a text box position of the adjusted driver's license image; a text recognition module configured to input the adjusted driver's license picture and the text box position into a text recognition network to obtain text information of the adjusted driver's license picture; and The text verification module is configured to verify the text information and output the text information that has been verified to be correct.

7. An electronic device, characterized in that: include: A processor; a memory storing executable instructions; Wherein, when the executable instruction is executed by the processor, the driver's license information recognition method as described in any one of claims 1-5 is implemented.

8. A computer-readable storage medium for storing a program, characterized in that: When the program is executed, the driver's license information recognition method as described in any one of claims 1-5 is implemented.

Images