License plate number fuzzy matching method and device and computer equipment

Abstract

The invention discloses a license plate number fuzzy matching method and device and computer equipment. The method comprises the following steps: when an entrance record of a vehicle needing to exit from a parking lot cannot be identified, acquiring first part license plate information and at least one piece of first feature information of the vehicle from a first video or picture of the vehicle captured by a camera, second videos or photos of vehicles covering the first part of license plate information are screened out from all vehicle entrance records of the parking lot, information reading is carried out on the second videos or photos, and second feature information of all the vehicles is obtained through reading; wherein the second feature information comprises feature information corresponding to all the first feature information, the second feature information is matched with the first feature information of the corresponding type, and the entering record of the vehicle matched with the second feature information with the highest priority in the matching result is determined as the entering record of the vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of parking lot management, and in particular, to a method and device for fuzzy matching of license plate numbers and a computer device. Background Art

[0002] Due to site reasons or vehicle - related reasons in some parking lots, when a vehicle exits the lot, it often fails to find the entry record, resulting in the inability to correctly match the entry record, and thus the vehicle cannot be charged normally when exiting the lot. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method and device for fuzzy matching of license plate numbers and a computer device, which can still match the entry record of a vehicle and charge the vehicle normally when the vehicle cannot correctly match the entry record when exiting the lot.

[0004] According to one aspect of the present invention, a method for fuzzy matching of license plate numbers is provided, including: when the entry record of a vehicle that needs to exit the parking lot cannot be recognized, obtaining the first - part license plate information of the vehicle and at least one first - feature information from the first video or photo of the vehicle captured by a camera; screening out the second video or photo of the vehicle covering the first - part license plate information from all the vehicle entry records in the parking lot; reading information from the second video or photo to obtain the second - feature information of each vehicle, where the second - feature information includes the feature information corresponding to all the first - feature information; matching the second - feature information with the first - feature information of the corresponding type, and determining the entry record of the vehicle matched by the second - feature information with the highest priority in the matching result as the entry record of the vehicle.

[0005] Among them, the step of, when the entry record of a vehicle that needs to exit the parking lot cannot be recognized, obtaining the first - part license plate information of the vehicle and at least one first - feature information from the first video or photo of the vehicle captured by a camera includes: when the entry record of a vehicle that needs to exit the parking lot cannot be recognized, extracting multiple videos or photos at different time points and different angles from the first video or photo of the vehicle captured by the camera, comparing the first videos or images at different time points and different perspectives, and combining the multi - source data fusion method to obtain the first - part license plate information of the vehicle and at least one first - feature information from the multiple videos or photos at different time points and different angles. The at least one first - feature information is the significant features of the vehicle, including vehicle type, color, unique logo on the vehicle body, vehicle - type contour, vehicle sticker, scratch, and special decoration.

[0006] Among them, screening out the second video or photo of the vehicle covering the first part of the license plate information from all the vehicle entry records of the parking lot includes: preprocessing all the vehicle entry records of the parking lot using a license plate recognition algorithm to extract license plate information, comparing the extracted license plate information with the first part of the license plate information, comparing the second license plate information covering the first part of the license plate information, and screening out the second video or photo of the vehicle associated with the second license plate information from all the vehicle entry records of the parking lot.

[0007] Among them, reading information from the second video or photo to obtain the second feature information of each vehicle; among them, the second feature information includes the feature information corresponding to all the first feature information, including: reading information from the second video or photo, using an image matching algorithm combined with deep learning technology to perform multi-image comparison on each vehicle image in the second video or photo, and reading the second feature information of each vehicle image including significant features such as vehicle type, color, unique markings on the vehicle body, vehicle type contour, car stickers, scratches, and special decorations, and using color recognition technology to read and classify the color of each vehicle, and using a deep learning algorithm to read and classify the vehicle type of each vehicle including sedans, sports utility vehicles, and sports cars. At the same time, during the process of reading the second feature information, comparing the read second feature information with the first feature information to ensure the coverage of the second feature information and the first feature information, so that the second feature information includes the feature information corresponding to all the first feature information.

[0008] Among them, matching the second feature information with the first feature information of the corresponding type, and determining the entry record of the vehicle matched by the second feature information with the highest priority in the matching result as the entry record of the vehicle includes: constructing a database containing various first feature information and associating the corresponding license plate information, preprocessing the second feature information to extract key features to ensure the same format as the first feature information in the database, and using a cosine similarity matching algorithm to match the similarity between the second feature information and the first feature information in the database. According to the similarity ranking, determining the entry record of the vehicle matched by the second feature information with the highest priority in the matching result as the entry record of the vehicle.

[0009] Among them, after matching the second feature information with the first feature information of the corresponding type and determining the entry record of the vehicle matched by the second feature information with the highest priority in the matching result as the entry record of the vehicle, it further includes: charging the parking record of the vehicle according to the entry record and notifying the payment.

[0010] According to another aspect of the present invention, there is provided a license plate number fuzzy matching device, comprising: an acquisition module, a screening module, a reading module and a determination module; the acquisition module is configured to, when the entry record of the vehicle that needs to exit the parking lot cannot be recognized, obtain the first part of the license plate information of the vehicle and at least one first feature information from the first video or photo of the vehicle captured by the camera; the screening module is configured to screen out the second video or photo of the vehicle covering the first part of the license plate information from all the vehicle entry records in the parking lot; the reading module is configured to read information from the second video or photo, and obtain the second feature information of each vehicle; wherein, the second feature information includes the feature information corresponding to all the first feature information; the determination module is configured to match the second feature information with the first feature information of the corresponding type, and determine the entry record of the vehicle matched by the second feature information with the highest priority in the matching result as the entry record of the vehicle.

[0011] Wherein, the acquisition module is specifically configured to: when the entry record of the vehicle that needs to exit the parking lot cannot be recognized, extract multiple videos or photos at different time points and different angles from the first video or photo of the vehicle captured by the camera, compare the first videos or images at different time points and different perspectives, and combine the multi-source data fusion method to obtain the first part of the license plate information of the vehicle and at least one first feature information from the multiple first videos or photos at different time points and different angles, wherein the at least one first feature information is the significant features of the vehicle, including vehicle type, color, unique identifier on the vehicle body, vehicle type contour, car sticker, scratch, special decoration.

[0012] Wherein, the screening module is specifically configured to: perform preprocessing on all the vehicle entry records in the parking lot by using a license plate recognition algorithm, extract license plate information, compare the extracted license plate information with the first part of the license plate information, compare out the second license plate information covering the first part of the license plate information, and screen out the second video or photo of the vehicle associated with the second license plate information from all the vehicle entry records in the parking lot.

[0013] Among them, the reading module is specifically configured to: read information from the second video or photo, use an image matching algorithm combined with deep learning technology to perform multi-image comparison on each vehicle image in the second video or photo, and read the second feature information of each vehicle image, including significant features such as vehicle model, color, unique identifier on the vehicle body, vehicle model contour, vehicle sticker, scratch, and special decoration, and use color recognition technology to read and classify the color of each vehicle, and use a deep learning algorithm to read and classify the vehicle models of each vehicle, including sedans, sports utility vehicles, and sports cars. At the same time, during the process of reading the second feature information, compare the read second feature information with the first feature information to ensure the coverage of the second feature information and the first feature information, so that the second feature information includes all the feature information corresponding to the first feature information.

[0014] Among them, the determination module is specifically configured to: construct a database containing various first feature information and associate the corresponding license plate information, preprocess the second feature information, extract key features to ensure that the format is consistent with the first feature information in the database, and use the cosine similarity matching algorithm to match the similarity between the second feature information and the first feature information in the database. According to the similarity ranking, determine the entry record of the vehicle with the highest priority second feature information match in the matching result as the entry record of the vehicle.

[0015] Among them, the license plate number fuzzy matching device further includes a charging module for charging the parking record of the vehicle according to the entry record and notifying the payment.

[0016] According to another aspect of the present invention, there is provided a computer device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the license plate number fuzzy matching method as described in any one of the above.

[0017] According to still another aspect of the present invention, there is provided a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the license plate number fuzzy matching method as described in any one of the above is implemented.

[0018] It can be found that in the above solution, when the entry record of the vehicle that needs to leave the parking lot cannot be recognized, the first part of the license plate information and at least one first feature information of the vehicle can be obtained from the first video or photo of the vehicle captured by the camera, and the second video or photo of the vehicle covering the first part of the license plate information can be screened out from all the vehicle entry records in the parking lot, and the information of the second video or photo can be read to obtain the second feature information of each vehicle; wherein, the second feature information includes the feature information corresponding to all the first feature information, and the second feature information is matched with the first feature information of the corresponding type, and the entry record of the vehicle matched by the second feature information with the highest priority in the matching result is determined as the entry record of the vehicle, which can realize matching the entry record of the vehicle and normally charging the vehicle exit fee when the vehicle cannot correctly match the entry record.

[0019] Furthermore, in the above solution, when the entry record of the vehicle that needs to leave the parking lot cannot be recognized, multiple videos or photos at different time points and different angles can be extracted from the first video or photo of the vehicle captured by the camera, and the first videos or images at different time points and different perspectives can be compared, and in combination with the multi-source data fusion method, the first part of the license plate information and at least one first feature information of the vehicle can be obtained from the multiple first videos or photos at different time points and different angles, wherein the at least one first feature information is the significant features of the vehicle, including vehicle type, color, unique markings on the vehicle body, vehicle type contour, car stickers, scratches, special decorations, etc. The advantage of this is to extract rich vehicle image data, enhance the adaptability to the complex environment of the parking lot, and can realize the accurate and efficient recognition of vehicle information.

[0020] Furthermore, in the above solution, a license plate recognition algorithm can be used to preprocess all the vehicle entry records in the parking lot, extract the license plate information, compare the extracted license plate information with the first part of the license plate information, compare the second license plate information covering the first part of the license plate information, and screen out the second video or photo of the vehicle associated with the second license plate information from all the vehicle entry records in the parking lot. The advantage of this is that through license plate information extraction, it can realize efficient and accurate matching of the second license plate information covering the first part of the license plate information, and then can realize efficient and accurate screening of the second video or photo of the vehicle associated with the second license plate information.

[0021] Furthermore, in the above solution, information can be read from the second video or photo. By using an image matching algorithm combined with deep learning technology, multi-image comparison is performed on each vehicle image in the second video or photo, and the second feature information of each vehicle image is read, including significant features such as vehicle type, color, unique markings on the vehicle body, vehicle type contour, decals, scratches, special decorations, etc. Also, by using color recognition technology, the colors of each vehicle are read and classified, and deep learning algorithms are used to read and classify the vehicle types of each vehicle, including sedans, sport utility vehicles, sports cars, etc. Meanwhile, during the process of reading the second feature information, the read second feature information is compared with the first feature information to ensure the coverage of the second feature information and the first feature information, so that the second feature information includes all the feature information corresponding to the first feature information. The advantage of this is that, due to the high adaptability and robustness of the image matching algorithm combined with deep learning technology, color recognition technology, and deep learning algorithms, etc., it can effectively improve the accuracy and efficiency of reading the second feature information of each vehicle.

[0022] Furthermore, in the above solution, a database containing various first feature information can be constructed, and the corresponding license plate information can be associated. Also, preprocessing can be performed on the second feature information to extract key features to ensure consistency with the format of the first feature information in the database. Then, matching algorithms such as cosine similarity are used to match the similarity between the second feature information and the first feature information in the database. According to the similarity ranking, the entry record of the vehicle with the highest priority second feature information match is determined as the entry record of the vehicle, which can achieve matching the entry record of the vehicle and correctly charging the vehicle exit fee even when the vehicle cannot correctly match the entry record.

[0023] Furthermore, in the above solution, the parking record of the vehicle can be charged according to the entry record and a payment notice can be sent. The advantage of this is that it can achieve matching the entry record of the vehicle and correctly charging the vehicle exit fee even when the vehicle cannot correctly match the entry record. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0025] Figure 1 It is a flowchart of an embodiment of the license plate number fuzzy matching method of the present invention;

[0026] Figure 2It is a schematic flowchart of another embodiment of the license plate number fuzzy matching method of the present invention;

[0027] Figure 3 It is a schematic structural diagram of an embodiment of the license plate number fuzzy matching device of the present invention;

[0028] Figure 4 It is a schematic structural diagram of another embodiment of the license plate number fuzzy matching device of the present invention;

[0029] Figure 5 It is a schematic structural diagram of an embodiment of the computer device of the present invention. Detailed implementation manners

[0030] The present invention will be further described in detail below in conjunction with the drawings and embodiments. It should be specifically noted that the following embodiments are only used to illustrate the present invention, but do not limit the scope of the present invention. Similarly, the following embodiments are only partial embodiments of the present invention rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0031] The present invention provides a license plate number fuzzy matching method, which can realize matching the entry record of a vehicle and the exit fee of a normally billed vehicle even when the vehicle cannot correctly match the entry record.

[0032] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of an embodiment of the license plate number fuzzy matching method of the present invention. It should be noted that if there are substantially the same results, the method of the present invention is not limited to Figure 1 the process sequence shown. As Figure 1 shown, the method includes the following steps:

[0033] S101: When the entry record of the vehicle that needs to exit the parking lot cannot be recognized, obtain the first part of the license plate information and at least one first feature information of the vehicle from the first video or photo of the vehicle captured by the camera.

[0034] Among them, when the entry record of the vehicle that needs to exit the parking lot cannot be recognized, obtaining the first part of the license plate information and at least one first feature information of the vehicle from the first video or photo of the vehicle captured by the camera may include:

[0035] When the entry record of the vehicle that needs to exit the parking lot cannot be recognized, multiple videos or photos at different time points and different angles are extracted from the first video or photo of the vehicle captured by the camera, and the first videos or images at different time points and different perspectives are compared. By combining the method of multi-source data fusion, the first partial license plate information of the vehicle and at least one first feature information are obtained from the multiple first videos or photos at different time points and different angles. The at least one first feature information is the significant features of the vehicle, including vehicle type, color, unique logo on the vehicle body, vehicle type contour, decal, scratch, special decoration, etc. The advantage is to extract rich vehicle image data, enhance the adaptability to the complex environment of the parking lot, and enable accurate and efficient identification of vehicle information.

[0036] S102: Screen out the second video or photo of the vehicle covering the first partial license plate information from all the vehicle entry records in the parking lot.

[0037] Among them, screening out the second video or photo of the vehicle covering the first partial license plate information from all the vehicle entry records in the parking lot may include:

[0038] Perform preprocessing on all the vehicle entry records in the parking lot using a license plate recognition algorithm, extract the license plate information, compare the extracted license plate information with the first partial license plate information, compare out the second license plate information covering the first partial license plate information, and screen out the second video or photo of the vehicle associated with the second license plate information from all the vehicle entry records in the parking lot. The advantage is that through license plate information extraction, it is possible to achieve efficient and accurate matching of the second license plate information covering the first partial license plate information, and then it is possible to achieve efficient and accurate screening of the second video or photo of the vehicle associated with the second license plate information.

[0039] S103: Read the information of the second video or photo to obtain the second feature information of each vehicle; among them, the second feature information includes the feature information corresponding to all the first feature information.

[0040] Among them, reading the information of the second video or photo to obtain the second feature information of each vehicle; among them, the second feature information includes the feature information corresponding to all the first feature information may include:

[0041] Read information from the second video or photo. Using an image matching algorithm combined with deep learning technology, perform multi-image comparison on each vehicle image in the second video or photo, and read the second feature information of each vehicle image, including significant features such as vehicle type, color, unique markings on the vehicle body, vehicle type contour, decals, scratches, special decorations, etc. And using color recognition technology, read and classify the colors of each vehicle, and use deep learning algorithms to read and classify the vehicle types of each vehicle, including sedans, SUVs (sport utility vehicles), sports cars, etc. At the same time, during the process of reading the second feature information, compare the read second feature information with the first feature information to ensure the coverage of the second feature information and the first feature information, so that the second feature information includes all the feature information corresponding to the first feature information. The advantage of this is that through the high adaptability and robustness of the image matching algorithm combined with deep learning technology, color recognition technology, and deep learning algorithms, etc., it can effectively improve the accuracy and efficiency of reading the second feature information of each vehicle.

[0042] S104: Match the second feature information with the first feature information of the corresponding type, and determine the entry record of the vehicle matched by the second feature information with the highest priority in the matching result as the entry record of the vehicle.

[0043] Among them, the step of matching the second feature information with the first feature information of the corresponding type and determining the entry record of the vehicle matched by the second feature information with the highest priority in the matching result as the entry record of the vehicle may include:

[0044] Construct a database containing various first feature information and associate the corresponding license plate information, preprocess the second feature information, extract key features to ensure the same format as the first feature information in the database, and use matching algorithms such as cosine similarity to match the similarity between the second feature information and the first feature information in the database. Sort according to the similarity, and determine the entry record of the vehicle matched by the second feature information with the highest priority in the matching result as the entry record of the vehicle, which can achieve matching the entry record of the vehicle and correctly charging the vehicle exit fee even when the vehicle cannot correctly match the entry record.

[0045] Among them, after the step of matching the second feature information with the first feature information of the corresponding type and determining the entry record of the vehicle matched by the second feature information with the highest priority in the matching result as the entry record of the vehicle, it may further include:

[0046] Billing the parking record of the vehicle and notifying payment according to the entry record. The advantage of this is that when the vehicle cannot be correctly matched with the entry record, the entry record of the vehicle and the normal charging of the vehicle's exit fee can still be matched.

[0047] It can be found that in this embodiment, when the entry record of the vehicle that needs to exit the parking lot cannot be recognized, the first partial license plate information and at least one first feature information of the vehicle can be obtained from the first video or photo of the vehicle captured by the camera, and the second video or photo of the vehicle covering the first partial license plate information can be screened out from all the vehicle entry records in the parking lot, and the second feature information of each vehicle can be read from the second video or photo; wherein, the second feature information includes the feature information corresponding to all the first feature information, and the second feature information is matched with the first feature information of the corresponding type, and the entry record of the vehicle matched with the second feature information with the highest priority in the matching result is determined as the entry record of the vehicle, so that when the vehicle cannot be correctly matched with the entry record, the entry record of the vehicle and the normal charging of the vehicle's exit fee can still be matched.

[0048] Furthermore, in this embodiment, when the entry record of the vehicle that needs to exit the parking lot cannot be recognized, multiple videos or photos at different time points and different angles can be extracted from the first video or photo of the vehicle captured by the camera, and the first videos or images at different time points and different perspectives can be compared, and in combination with the multi-source data fusion method, the first partial license plate information and at least one first feature information of the vehicle can be obtained from the multiple videos or photos at different time points and different angles. Among them, the at least one first feature information is the significant feature of the vehicle, including vehicle type, color, unique identifier on the vehicle body, vehicle type contour, car sticker, scratch, special decoration, etc. The advantage of this is to extract rich vehicle image data, enhance the adaptability to the complex environment of the parking lot, and enable accurate and efficient identification of vehicle information.

[0049] Furthermore, in this embodiment, a license plate recognition algorithm can be used to preprocess all the vehicle entry records in the parking lot, extract license plate information, and compare the extracted license plate information with the first partial license plate information to compare and obtain the second license plate information covering the first partial license plate information, and screen out the second video or photo of the vehicle associated with the second license plate information from all the vehicle entry records in the parking lot. The advantage of this is that through license plate information extraction, the second license plate information covering the first partial license plate information can be efficiently and accurately matched, and then the second video or photo of the vehicle associated with the second license plate information can be efficiently and accurately screened out.

[0050] Further, in this embodiment, information can be read from the second video or photo. By using an image matching algorithm combined with deep learning technology, multi-image comparison is performed on each vehicle image in the second video or photo. The second feature information read from each vehicle image includes significant features such as vehicle model, color, unique markings on the vehicle body, vehicle model outline, decals, scratches, special decorations, etc. And by using color recognition technology, the color of each vehicle is read and classified, and deep learning algorithms are used to read and classify the vehicle models of each vehicle, including sedans, sport utility vehicles, sports cars, etc. At the same time, during the process of reading the second feature information, the read second feature information is compared with the first feature information to ensure the coverage of the second feature information and the first feature information, so that the second feature information includes all the feature information corresponding to the first feature information. The advantage of this is that through the high adaptability and robustness of the image matching algorithm combined with deep learning technology, color recognition technology, and deep learning algorithms, etc., it can effectively improve the accuracy and efficiency of reading the second feature information of each vehicle.

[0051] Further, in this embodiment, a database containing various first feature information can be constructed, and the corresponding license plate information can be associated. And the second feature information is preprocessed to extract key features to ensure that the format is consistent with the first feature information in the database. And matching algorithms such as cosine similarity are used to match the similarity between the second feature information and the first feature information in the database. According to the similarity ranking, the entry record of the vehicle with the highest priority second feature information match is determined as the entry record of the vehicle, which can achieve matching the entry record of the vehicle and the normal charging of the vehicle's exit fee even when the vehicle cannot correctly match the entry record.

[0052] Please refer to Figure 2 , Figure 2 is a flowchart of another embodiment of the license plate number fuzzy matching method of the present invention. In this embodiment, the method includes the following steps:

[0053] S201: When the entry record of the vehicle that needs to exit the parking lot cannot be recognized, obtain the first part of the license plate information and at least one first feature information of the vehicle from the first video or photo of the vehicle captured by the camera.

[0054] It can be as described in S101 above and will not be elaborated here.

[0055] S202: Screen out the second video or photo of the vehicle that covers the first part of the license plate information from all the vehicle entry records in the parking lot.

[0056] It can be as described in S102 above and will not be elaborated here.

[0057] S203: Read information from the second video or photo to obtain the second feature information of each vehicle; wherein, the second feature information includes the feature information corresponding to all the first feature information.

[0058] As described in S103 above, details are not elaborated here.

[0059] S204: Match the second feature information with the first feature information of the corresponding type, and determine the entry record of the vehicle with the highest-priority second feature information in the matching result as the entry record of the vehicle.

[0060] As described in S104 above, details are not elaborated here.

[0061] S205: Charge the parking record of the vehicle according to the entry record and notify for payment.

[0062] It can be found that in this embodiment, the parking record of the vehicle can be charged according to the entry record and notified for payment. The advantage of this is that it can achieve matching the entry record of the vehicle and normally charging the vehicle exit fee even when the vehicle cannot correctly match the entry record.

[0063] The present invention also provides a license plate number fuzzy matching device, which can achieve matching the entry record of the vehicle and normally charging the vehicle exit fee even when the vehicle cannot correctly match the entry record.

[0064] Please refer to Figure 3 , Figure 3 which is a schematic structural diagram of an embodiment of the license plate number fuzzy matching device of the present invention. In this embodiment, the license plate number fuzzy matching device 30 includes an acquisition module 31, a screening module 32, a reading module 33, and a determination module 34.

[0065] The acquisition module 31 is configured to obtain the first partial license plate information and at least one first feature information of the vehicle from the first video or photo of the vehicle captured by the camera when the entry record of the vehicle that needs to exit the parking lot cannot be recognized.

[0066] The screening module 32 is configured to screen out the second video or photo of the vehicle covering the first partial license plate information from all the vehicle entry records in the parking lot.

[0067] The reading module 33 is configured to read information from the second video or photo to obtain the second feature information of each vehicle; wherein, the second feature information includes the feature information corresponding to all the first feature information.

[0068] The determination module 34 is configured to match the second feature information with the first feature information of the corresponding type, and determine the entry record of the vehicle matched by the second feature information with the highest priority in the matching result as the entry record of the vehicle.

[0069] Optionally, the obtaining module 31 may be specifically configured to:

[0070] When the entry record of the vehicle that needs to exit the parking lot cannot be recognized, multiple videos or photos at different time points and different angles are extracted from the first video or photo of the vehicle captured by the camera, and the first videos or images at different time points and different perspectives are compared, and in combination with the multi-source data fusion method, the first partial license plate information and at least one first feature information of the vehicle are obtained from the multiple first videos or photos at different time points and different angles, where the at least one first feature information is the significant features of the vehicle, including vehicle type, color, unique identifier on the vehicle body, vehicle type contour, car sticker, scratch, special decoration, etc.

[0071] Optionally, the screening module 32 may be specifically configured to:

[0072] Preprocess all vehicle entry records in the parking lot using a license plate recognition algorithm, extract the license plate information, compare the extracted license plate information with the first partial license plate information, compare out the second license plate information covering the first partial license plate information, and screen out the second video or photo of the vehicle associated with the second license plate information from all vehicle entry records in the parking lot.

[0073] Optionally, the reading module 33 may be specifically configured to:

[0074] Read the information of the second video or photo, use the image matching algorithm combined with deep learning technology to perform multi-image comparison on each vehicle image in the second video or photo, read out the second feature information of each vehicle image, including significant features such as vehicle type, color, unique identifier on the vehicle body, vehicle type contour, car sticker, scratch, special decoration, etc., and use the color recognition technology to read and classify the color of each vehicle, and use the deep learning algorithm to read and classify the vehicle type of each vehicle, including sedans, sports utility vehicles, sports cars, etc. At the same time, during the process of reading the second feature information, compare the read second feature information with the first feature information to ensure the coverage of the second feature information and the first feature information, so that the second feature information includes all the feature information corresponding to the first feature information.

[0075] Optionally, the determination module 34 may be specifically configured to:

[0076] Construct a database containing various first feature information, associate the corresponding license plate information, preprocess the second feature information, extract key features to ensure they are in the same format as the first feature information in the database, and use matching algorithms such as cosine similarity to match the similarity between the second feature information and the first feature information in the database. Sort according to the similarity, and determine the entry record of the vehicle with the highest-priority second feature information in the matching result as the entry record of the vehicle.

[0077] Please refer to Figure 4 , Figure 4 FIG. is a schematic structural diagram of another embodiment of the license plate number fuzzy matching device of the present invention. Different from the previous embodiment, the license plate number fuzzy matching device 40 in this embodiment further includes a charging module 41.

[0078] The charging module 41 is used to charge the parking record of the vehicle according to the entry record and notify the payment.

[0079] Each unit module of the license plate number fuzzy matching device 30 / 40 can respectively execute the corresponding steps in the above method embodiments, so the unit modules will not be described in detail here. For details, please refer to the description of the corresponding steps above.

[0080] The present invention also provides a computer device, as Figure 5 shown, including: at least one processor 51; and a memory 52 communicatively connected to the at least one processor 51; wherein, the memory 52 stores instructions executable by the at least one processor 51, and the instructions are executed by the at least one processor 51 to enable the at least one processor 51 to execute the above license plate number fuzzy matching method.

[0081] Among them, the memory 52 and the processor 51 are connected by a bus. The bus can include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors 51 and the memory 52 together. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits together, which are well known in the art. Therefore, they will not be further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be an element or multiple elements, such as multiple receivers and transmitters, and provides a unit for communicating with various other devices on the transmission medium. The data processed by the processor 51 is transmitted on the wireless medium through the antenna. Further, the antenna also receives data and transmits the data to the processor 51.

[0082] The processor 51 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. And the memory 52 can be used to store the data used by the processor 51 when executing operations.

[0083] The present invention further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the above method embodiments are implemented.

[0084] It can be found that in the above solution, when the entry record of the vehicle that needs to exit the parking lot cannot be recognized, the first partial license plate information and at least one first feature information of the vehicle can be obtained from the first video or photo of the vehicle captured by the camera, and the second video or photo of the vehicle covering the first partial license plate information can be screened out from all the vehicle entry records of the parking lot, and information can be read from the second video or photo to obtain the second feature information of each vehicle; wherein, the second feature information includes the feature information corresponding to all the first feature information, and the second feature information is matched with the first feature information of the corresponding type, and the entry record of the vehicle matched with the second feature information with the highest priority in the matching result is determined as the entry record of the vehicle, so that it can be realized that when the vehicle cannot correctly match the entry record, the entry record of the vehicle can still be matched and the exit fee of the vehicle can be normally charged.

[0085] Furthermore, in the above solution, when the entry record of the vehicle that needs to exit the parking lot cannot be recognized, multiple videos or photos at different time points and different angles can be extracted from the first video or photo of the vehicle captured by the camera, and the first videos or images at different time points and different perspectives can be compared, and in combination with the multi-source data fusion method, the first partial license plate information and at least one first feature information of the vehicle can be obtained from the multiple videos or photos at different time points and different angles, wherein the at least one first feature information is the significant feature of the vehicle, including vehicle type, color, unique identifier on the vehicle body, vehicle type contour, car sticker, scratch, special decoration, etc. The advantage of this is to extract rich vehicle image data, enhance the adaptability to the complex environment of the parking lot, and be able to realize the accurate and efficient recognition of vehicle information.

[0086] Furthermore, in the above solution, a license plate recognition algorithm can be used to preprocess all the vehicle entry records of the parking lot to extract license plate information, and the extracted license plate information is compared with the first partial license plate information to compare and obtain the second license plate information covering the first partial license plate information, and the second video or photo of the vehicle associated with the second license plate information is screened out from all the vehicle entry records of the parking lot. The advantage of this is that through license plate information extraction, it can be realized to efficiently and accurately match the second license plate information covering the first partial license plate information, and further can be realized to efficiently and accurately screen out the second video or photo of the vehicle associated with the second license plate information.

[0087] Furthermore, in the above solution, information can be read from the second video or photo. By using an image matching algorithm combined with deep learning technology, multi-image comparison is performed on each vehicle image in the second video or photo, and the second feature information of each vehicle image is read, including significant features such as vehicle model, color, unique markings on the vehicle body, vehicle model outline, decals, scratches, special decorations, etc. Also, by using color recognition technology, the color of each vehicle is read and classified, and by using a deep learning algorithm, the vehicle models including sedans, sport utility vehicles, sports cars, etc. are read and classified. Meanwhile, during the process of reading the second feature information, the read second feature information is compared with the first feature information to ensure the coverage of the second feature information and the first feature information, so that the second feature information includes all the feature information corresponding to the first feature information. The advantage of this is that with the high adaptability and robustness of the image matching algorithm combined with deep learning technology, color recognition technology, and deep learning algorithm, etc., the accuracy and efficiency of reading the second feature information of each vehicle can be effectively improved.

[0088] Furthermore, in the above solution, a database containing various first feature information can be constructed, and the corresponding license plate information can be associated. Also, the second feature information can be preprocessed to extract key features to ensure consistency with the format of the first feature information in the database. Then, by using matching algorithms such as cosine similarity, the similarity between the second feature information and the first feature information in the database is matched. According to the similarity ranking, the entry record of the vehicle with the highest priority second feature information match is determined as the entry record of the vehicle, which can achieve matching the entry record of the vehicle and correctly charging the vehicle's exit fee even when the vehicle cannot correctly match the entry record.

[0089] Furthermore, in the above solution, the parking record of the vehicle can be charged according to the entry record and the payment can be notified. The advantage of this is that it can achieve matching the entry record of the vehicle and correctly charging the vehicle's exit fee even when the vehicle cannot correctly match the entry record.

[0090] In several implementation manners provided by the present invention, it should be understood that the disclosed system, device, and method can be implemented in other ways. For example, the device implementation manner described above is only illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be in electrical, mechanical, or other forms.

[0091] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0092] In addition, each functional unit in various embodiments of the present invention can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0093] If the integrated unit is implemented in the form of 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, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods in various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs and other various media that can store program codes.

[0094] The above are only partial embodiments of the present invention and do not limit the protection scope of the present invention accordingly. Any equivalent device or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be similarly included in the patent protection scope of the present invention.

Claims

1. A license plate number fuzzy matching method, characterized in that: include: When an entry record of a vehicle that needs to exit the parking lot cannot be identified, obtaining a first portion of license plate information and at least one first feature information of the vehicle from a first video or photo of the vehicle captured by a camera; Filtering a second video or photo of a vehicle covering the first part of the license plate information from all vehicle entry records of the parking lot; Reading the second video or photo to obtain second characteristic information of each vehicle; wherein the second characteristic information includes characteristic information corresponding to all the first characteristic information; The second characteristic information is matched with the first characteristic information of the corresponding type, and the entry record of the vehicle matched with the second characteristic information with the highest priority in the matching results is determined as the entry record of the vehicle.

2. The license plate number fuzzy matching method according to claim 1, characterized in that: When the entry record of the vehicle that needs to exit the parking lot cannot be identified, obtaining the first part of the license plate information and at least one first feature information of the vehicle from the first video or photo of the vehicle captured by the camera includes: When the entry record of the vehicle that needs to exit the parking lot cannot be identified, multiple videos or photos taken at different time points and angles are extracted from the first video or photo of the vehicle captured by the camera, and the first videos or images at different time points and different perspectives are compared, and combined with multi-source data fusion, the first part of the license plate information and at least one first feature information of the vehicle are obtained from the multiple first videos or photos taken at different time points and angles, wherein the at least one first feature information is a significant feature of the vehicle including vehicle model, color, unique logo on the body, vehicle model outline, vehicle sticker, scratches, and special decorations.

3. The license plate number fuzzy matching method according to claim 1, characterized in that: The step of selecting a second video or photo of a vehicle covering the first part of the license plate information from all vehicle entry records of the parking lot includes: A license plate recognition algorithm is used to pre-process all vehicle entry records of the parking lot, extract license plate information, and compare the extracted license plate information with the first part of the license plate information to obtain second license plate information covering the first part of the license plate information, and filter out a second video or photo of the vehicle associated with the second license plate information from all vehicle entry records of the parking lot.

4. The license plate number fuzzy matching method according to claim 1, characterized in that: The second video or photo is read to obtain second characteristic information of each vehicle; wherein the second characteristic information includes characteristic information corresponding to all the first characteristic information, including: The information of the second video or photo is read, and multiple images of each vehicle image in the second video or photo are compared by using an image matching algorithm combined with a deep learning technology. The second feature information of each vehicle image is read out, including significant features such as vehicle model, color, unique mark on the vehicle body, vehicle model outline, vehicle sticker, scratches, special decorations, and color recognition technology is used to read and classify the color of each vehicle, and a deep learning algorithm is used to read and classify the vehicle model of each vehicle including sedans, sports utility vehicles, and sports cars. At the same time, in the process of reading the second feature information, the read second feature information is compared with the first feature information to ensure the coverage of the second feature information and the first feature information, so that the second feature information includes all feature information corresponding to the first feature information.

5. The license plate number fuzzy matching method according to claim 1, characterized in that: The matching of the second characteristic information with the first characteristic information of the corresponding type, and determining the entry record of the vehicle matched with the second characteristic information with the highest priority in the matching results as the entry record of the vehicle, includes: Construct a database containing multiple first feature information, and associate the corresponding license plate information, and pre-process the second feature information to extract key features to ensure consistency with the format of the first feature information in the database, and use a cosine similarity matching algorithm to match the second feature information with the similarity of the first feature information in the database, sort according to the similarity, and determine the entry record of the vehicle matched by the second feature information with the highest priority in the matching results as the entry record of the vehicle.

6. The license plate number fuzzy matching method according to claim 1, characterized in that: After matching the second characteristic information with the first characteristic information of the corresponding type and determining the entry record of the vehicle matched with the second characteristic information with the highest priority in the matching results as the entry record of the vehicle, the method further includes: The parking record of the vehicle is charged according to the entry record and payment is notified.

7. A license plate number fuzzy matching device, characterized in that: include: an acquisition module, a screening module, a reading module and a determination module; The acquisition module is used to acquire the first part of the license plate information and at least one first feature information of the vehicle from the first video or photo of the vehicle captured by the camera when the entry record of the vehicle that needs to exit the parking lot cannot be identified; The screening module is used to screen out the second video or photo of the vehicle covering the first part of the license plate information from all the vehicle entry records of the parking lot; The reading module is used to read the second video or photo to obtain the second characteristic information of each vehicle; wherein the second characteristic information includes characteristic information corresponding to all the first characteristic information; The determination module is used to match the second feature information with the first feature information of the corresponding type, and determine the entry record of the vehicle matched by the second feature information with the highest priority in the matching results as the entry record of the vehicle.

8. The vehicle license plate number fuzzy matching device as claimed in claim 7, characterized in that: The acquisition module is specifically used for: When the entry record of the vehicle that needs to exit the parking lot cannot be identified, multiple videos or photos taken at different time points and angles are extracted from the first video or photo of the vehicle captured by the camera, and the first videos or images at different time points and different perspectives are compared, and combined with multi-source data fusion, the first part of the license plate information and at least one first feature information of the vehicle are obtained from the multiple first videos or photos taken at different time points and angles, wherein the at least one first feature information is a significant feature of the vehicle including vehicle model, color, unique logo on the body, vehicle model outline, vehicle sticker, scratches, and special decorations.

9. A computer device, characterized in that: include: at least one processor; And, a memory communicatively connected to at least one processor; wherein the memory stores instructions executable by at least one processor, and the instructions are executed by at least one processor so that at least one processor can execute the license plate number fuzzy matching method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the license plate number fuzzy matching method as described in any one of claims 1 to 6 is implemented.

Images