Vehicle management and control method and system, computer equipment and storage medium
Through video data matching technology, the problem of handling unlicensed vehicles and vehicles without entry records is solved, and the rapid release of vehicles without license plates and entry records is achieved, which improves the accuracy and user experience of parking lot management.
Patent Information
- Application Number
- CN202411009239.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-11-22
AI Technical Summary
The prior art cannot accurately and quickly deal with unlicensed vehicles and vehicles without entry records, resulting in poor user experience and loss of parking lot revenue.
By obtaining vehicle information in the video data and re-identifying feature vectors, the target vehicle is matched with the unmatched successful entry vehicle, and the similarity is calculated to determine the entry record, so as to achieve rapid release of vehicles without license plates and entry records.
It improves user experience, improves the accuracy of parking lot control, and reduces the revenue loss of parking lots.
Smart Images

Figure CN119007454A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent video surveillance technology, and in particular to a vehicle control method and system, computer equipment and storage medium. Background Art
[0002] Unattended parking is a convenient parking lot management mode that minimizes human intervention while ensuring the safety of vehicles and property, and can provide efficient services such as touchless parking, online payment, intelligent reminders, and remote calls. In order to facilitate the control of vehicles entering and leaving the parking lot, vehicle identification equipment is usually installed at the entrance and exit of the parking lot to perform operations such as license plate recognition on vehicles entering and leaving the parking lot, so as to achieve rapid entry and exit and control of vehicles.
[0003] However, due to the recognition strategy configuration of the vehicle identification equipment, normal exit processing can be achieved for vehicles with license plates and entry records. However, for some special vehicles, such as vehicles without license plates and no entry records, manual participation in the vehicle entry and exit records and entry and exit control is still required, and accurate payment and rapid entry and exit release cannot be achieved, resulting in poor user experience and easily causing loss of parking lot income or increased labor costs. Summary of the invention
[0004] Based on this, it is necessary to provide a vehicle control method, system, computer device and storage medium to address the above technical issues, so as to solve at least one problem existing in the above-mentioned prior art.
[0005] The embodiment of the present application is implemented as follows. In a first aspect, a vehicle control method is provided, comprising the following steps:
[0006] Obtain vehicle information and vehicle re-identification feature vectors of all vehicles entering the area to be identified in the video data;
[0007] When the target vehicle exits, the vehicle re-identification feature vector and vehicle information of the target vehicle are obtained;
[0008] Respectively matching the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle to obtain a first target entry vehicle that matches the target vehicle;
[0009] An entry record corresponding to the first target entry vehicle is obtained, and based on the entry record corresponding to the first target entry vehicle, the target vehicle is released.
[0010] In one embodiment, the step of matching the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle to obtain a first target entry vehicle that matches the target vehicle includes:
[0011] Calculating a first similarity between the vehicle re-identification feature vector of the target vehicle and the vehicle re-identification feature vectors of each unsuccessfully matched entry vehicle;
[0012] Calculating a second similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle;
[0013] An unsuccessfully matched entry vehicle whose first similarity is greater than a first preset threshold and whose second similarity is greater than a second preset threshold is determined as the first target entry vehicle.
[0014] In one embodiment, after matching the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, the method further comprises:
[0015] When a first target entry vehicle matching the target vehicle is not obtained, weighted average values are taken for the first similarity and the second similarity to obtain a first overall similarity;
[0016] Sorting the admission records of the unmatched admission vehicles in descending order of the first overall similarity to obtain a first recommendation list;
[0017] The first recommendation list is recommended to parking lot management personnel.
[0018] In one embodiment, before matching the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, the process includes:
[0019] Determining whether the target vehicle is a licensed vehicle;
[0020] If yes, determining whether the target vehicle is a vehicle with an entry record;
[0021] If yes, the entry record of the target vehicle is obtained, and based on the entry record of the target vehicle, the target vehicle is released.
[0022] In one embodiment, after determining whether the target vehicle is a vehicle with an entry record, the following steps are performed:
[0023] If not, the target license plate, the vehicle re-identification feature vector and the vehicle information of the target vehicle are matched with the license plates, the vehicle re-identification feature vectors and the vehicle information of each of the unmatched entry vehicles to obtain a second target entry vehicle that matches the target vehicle;
[0024] An entry record corresponding to the second target entering vehicle is obtained, and based on the entry record corresponding to the second target entering vehicle, the target vehicle is released.
[0025] In one embodiment, the matching of the target license plate, the vehicle re-identification feature vector, and the vehicle information of the target vehicle with the license plate, the vehicle re-identification feature vector, and the vehicle information of each of the unmatched entry vehicles to obtain a second target entry vehicle that matches the target vehicle includes:
[0026] Calculating a third similarity between the target license plate of the target vehicle and the license plates of each of the unmatched entry vehicles;
[0027] Calculating a fourth similarity between the vehicle re-identification feature vector of the target vehicle and the vehicle re-identification feature vectors of each of the unsuccessfully matched entry vehicles;
[0028] Calculating a fifth similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle;
[0029] An unsuccessfully matched entry vehicle whose third similarity is greater than a third preset threshold, whose fourth similarity is greater than a fourth preset threshold, and whose fifth similarity is greater than a fifth preset threshold is determined as the second target entry vehicle.
[0030] In one embodiment, after matching the target license plate, the vehicle re-identification feature vector, and the vehicle information of the target vehicle with the license plate, the vehicle re-identification feature vector, and the vehicle information of each of the unmatched entry vehicles, the method further comprises:
[0031] When the second target entering vehicle is not obtained, weighted average values are respectively obtained for the third similarity, the fourth similarity and the fifth similarity to obtain a second overall similarity;
[0032] Sorting the admission records of the unmatched admission vehicles in descending order of the second overall similarity to obtain a second recommendation list;
[0033] The second recommendation list is recommended to parking lot management personnel.
[0034] In a second aspect, a vehicle control system is provided, comprising:
[0035] An entry vehicle data extraction unit, used to obtain vehicle information and vehicle re-identification feature vectors of all entry vehicles in the video data of the area to be identified;
[0036] A target vehicle data extraction unit is used to obtain a vehicle re-identification feature vector and vehicle information of the target vehicle when the target vehicle exits the scene;
[0037] A matching unit, used to match the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, so as to obtain a first target entry vehicle that matches the target vehicle;
[0038] The exit release unit is used to obtain the entry record corresponding to the first target entry vehicle, and release the target vehicle based on the entry record corresponding to the first target entry vehicle.
[0039] In a third aspect, a computer device is provided, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the steps of the vehicle control method as described above when executing the computer-readable instructions.
[0040] In a fourth aspect, a readable storage medium is provided, wherein the readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the vehicle control method as described above are implemented.
[0041] The above-mentioned vehicle control method, system, computer device and storage medium, the method implementation includes: obtaining vehicle information and vehicle re-identification feature vectors of all vehicles entering the area to be identified in the video data of the area to be identified; when the target vehicle exits the area, obtaining the vehicle re-identification feature vector and vehicle information of the target vehicle; matching the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entering vehicle, respectively, to obtain a first target entering vehicle that matches the target vehicle; obtaining the entry record corresponding to the first target entering vehicle, and releasing the target vehicle based on the entry record corresponding to the first target entering vehicle. In an embodiment of the present application, when a vehicle enters a parking lot, vehicle information is identified and a vehicle re-identification result is generated and stored. When a vehicle exits a parking lot, a vehicle without a license plate and without an entry record can be matched with the vehicle re-identification result and vehicle information of an unmatched entry vehicle to obtain an entry record with a similarity greater than a threshold, so as to calculate parking fees and control exit, thereby achieving rapid release of vehicles without license plates and / or entry records, thereby improving user experience, improving the accuracy of parking lot management and control, and reducing parking lot revenue losses. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.
[0043] Figure 1 A flow chart of a vehicle control method according to an embodiment of the present invention is shown in FIG. Figure 1 ;
[0044] Figure 2 A flow chart of a vehicle control method according to an embodiment of the present invention is shown in FIG. Figure 2 ;
[0045] Figure 3 A flow chart of a vehicle control method according to an embodiment of the present invention is shown in FIG. Figure 3 ;
[0046] Figure 4 A flow chart of a vehicle control method according to an embodiment of the present invention is shown in FIG. Figure 4 ;
[0047] Figure 5 A flow chart of a vehicle control method according to an embodiment of the present invention is shown in FIG. Figure 5 ;
[0048] Figure 6is a structural schematic diagram of a vehicle control system in one embodiment of the present invention;
[0049] Figure 7 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION
[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0051] In one embodiment, if Figure 1 As shown, a vehicle control method implementation process is provided, including the following steps:
[0052] In step S110, vehicle information and vehicle re-identification feature vectors of all vehicles entering the area to be identified are obtained in the video data of the area to be identified;
[0053] In an embodiment of the present application, the area to be identified may be the entrance area of the parking lot, which may include the outside of the entrance barrier, that is, within a preset range from the outside of the parking lot to the entrance barrier of the parking lot, and / or the inside of the entrance barrier, that is, within a preset range from the inside of the parking lot to the entrance barrier of the parking lot. Video data of the entire process of a vehicle entering the parking lot from the outside of the parking lot can be collected through pre-installed vehicle identification equipment, such as cameras.
[0054] It can be understood that the vehicle identification device may include multiple, for example, it may include two, which are respectively set at the entrance gate and face the inside and outside of the parking lot respectively, so as to capture the entire process of the vehicle driving from outside the parking lot into the parking lot.
[0055] In the embodiment of the present application, the vehicle information includes but is not limited to the license plate number, vehicle style, vehicle model, and vehicle color. Optionally, the video data can be split frame by frame to obtain continuous video frame images, and the vehicles appearing in each video frame are selected by a vehicle detection frame to extract the vehicle area, and a preset feature extraction model, such as a CNN model, is used to extract vehicle information such as the license plate number and vehicle color of the vehicle in the vehicle area.
[0056] In an embodiment of the present application, a re-identification model ReID can be constructed, and the re-identification model ReID can be trained and optimized by collecting sample data until it meets the convergence conditions. The collected video data is pre-processed and then input into the pre-trained re-identification model ReID for processing to obtain the re-identification results of each entering vehicle. The re-identification result may include a re-identification feature vector, and the re-identification feature vector and the vehicle information are stored in the vehicle entry record for subsequent use.
[0057] It should be noted that the vehicle information and vehicle re-identification feature vector can be identified and extracted by the vehicle identification device installed at the entrance. Alternatively, the parking management system can obtain the video data collected by the vehicle identification device, and identify and obtain the vehicle information and vehicle re-identification feature vector of the entering vehicle in the video data. The parking management system can be set locally or in the cloud.
[0058] In step S110, when the target vehicle exits, a vehicle re-identification feature vector and vehicle information of the target vehicle are obtained;
[0059] In an embodiment of the present application, video data of the parking lot exit area can be collected in real time, and the target vehicle in the video data can be detected by a target detection algorithm, such as the YOLO algorithm. When the target vehicle is detected and the target vehicle is about to exit, the vehicle re-identification feature vector and vehicle information of the target vehicle can be obtained.
[0060] Among them, the parking lot exit area may include the outside of the exit gate, that is, within a preset range from the outside of the parking lot to the parking lot exit gate, and / or the inside of the parking lot exit, that is, within a preset range from the inside of the parking lot to the parking lot exit gate. Pre-installed vehicle identification equipment, such as cameras, can be used to collect video data of the entire process of a vehicle leaving the parking lot from the inside of the parking lot to the outside of the parking lot.
[0061] It can be understood that the vehicle identification device may include multiple, for example, it may include two, which are respectively set at the exit gate and face the inside and outside of the parking lot respectively, so as to capture the entire process of the vehicle leaving the parking lot.
[0062] In the embodiment of the present application, the vehicle information of the target vehicle includes but is not limited to the license plate number, vehicle style, vehicle model, and vehicle color. Optionally, when the target vehicle exits, video data of the target vehicle in the exit area can be collected, and the video data can be split frame by frame to obtain continuous video frame images, and the vehicles appearing in each video frame are framed by a vehicle detection frame to extract the vehicle area, and a preset feature extraction model, such as a CNN model, is used to extract vehicle information such as the license plate number and vehicle color of the vehicle in the vehicle area.
[0063] In an embodiment of the present application, the collected video data may be pre-processed and then input into the pre-trained re-identification model ReID for processing to obtain a re-identification result of the target vehicle, which may include a vehicle re-identification feature vector.
[0064] It should be noted that the vehicle information and vehicle re-identification feature vector of the target vehicle can be identified and extracted by the vehicle identification device set at the exit. Alternatively, the parking management system can obtain the video data collected by the vehicle identification device, and identify and obtain the vehicle information and vehicle re-identification feature vector of the target vehicle in the video data. The parking management system can be set locally or in the cloud.
[0065] In step S130, the vehicle re-identification feature vector and vehicle information of the target vehicle are matched with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, so as to obtain a first target entry vehicle that matches the target vehicle;
[0066] In the embodiment of the present application, after obtaining the vehicle re-identification feature vector and vehicle information of the target vehicle, the entry record of each unmatched entry vehicle is retrieved, that is, the entry record of the entry vehicle that is not currently matched, so as to obtain the vehicle re-identification feature vector and vehicle information extracted when each unmatched entry vehicle enters the venue according to the entry record, and match them with the vehicle feature vector and vehicle information of the target vehicle respectively. Optionally, the similarity between the vehicle re-identification feature vector of the target vehicle and the re-identification feature vector of each unmatched entry vehicle, as well as the similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle can be calculated respectively by a cosine similarity algorithm, a Euclidean distance algorithm, etc.
[0067] It can be understood that the similarity between the vehicle re-identification feature vector of the target vehicle and the re-identification feature vector of each unmatched entering vehicle can be calculated first, and when the similarity reaches a preset threshold, the similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entering vehicle is calculated. Alternatively, the similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entering vehicle is calculated first, and when the similarity reaches a preset threshold, the similarity between the vehicle re-identification feature vector of the target vehicle and the re-identification feature vector of each unmatched entering vehicle is calculated, so as to save the calculation process and improve the matching speed.
[0068] It should be noted that the unmatched entry vehicle refers to the entry vehicle that has not been matched or failed to match when the target vehicle exits among all the collected entry vehicles. The first target entry vehicle refers to the entry vehicle that matches the target vehicle among all the unmatched entry vehicles.
[0069] In step S150, the entry record of the first target entering vehicle is obtained, and based on the entry record of the first target entering vehicle, the target vehicle is released.
[0070] In the embodiment of the present application, when the vehicle re-identification feature vector of the target vehicle successfully matches the vehicle re-identification feature vector of each unmatched entry vehicle, and the vehicle information of the target vehicle also successfully matches the vehicle information of each unmatched entry vehicle, the entry record of each unmatched entry vehicle can be used as the entry record of the target vehicle, so that the entry time of the target vehicle can be obtained in order to calculate the parking fee of the target vehicle. When the parking fee is paid, the target vehicle can be released. When the vehicle enters the parking lot, the re-identification vector feature and vehicle information of the vehicle are obtained, and when the vehicle exits the parking lot, the re-identification vector feature and vehicle information of the vehicle are obtained again, and matched with the vehicle re-identification feature vector and vehicle information obtained when entering the parking lot, so as to control the entry and exit of vehicles without license plates and vehicles without entry records, improve the accuracy and efficiency of parking lot management, and improve user experience.
[0071] In an embodiment of the present application, a vehicle control method is provided, including: obtaining vehicle information and vehicle re-identification feature vectors of all vehicles entering the parking lot in the video data of the area to be identified; when the target vehicle exits the parking lot, obtaining the vehicle re-identification feature vector and vehicle information of the target vehicle; matching the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched vehicle entering the parking lot, respectively, to obtain a first target vehicle entering the parking lot that matches the target vehicle; obtaining the entry record corresponding to the first target vehicle entering the parking lot, and releasing the target vehicle based on the entry record corresponding to the first target vehicle entering the parking lot. In an embodiment of the present application, by identifying the vehicle information when the vehicle enters the parking lot, and generating a vehicle re-identification result for storage, when the vehicle exits the parking lot, for a vehicle without a license plate and without an entry record, the vehicle re-identification result and vehicle information of the unmatched vehicle entering the parking lot can be matched to obtain an entry record with a similarity greater than a threshold, so as to calculate the cost of the vehicle and control the exit of the vehicle, thereby realizing the rapid release of vehicles without a license plate and / or an entry record, which can improve the user experience, improve the accuracy of parking lot control, and reduce parking lot revenue losses.
[0072] See also Figure 2 In one embodiment of the present application, the vehicle re-identification feature vector and vehicle information of the target vehicle are matched with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle to obtain a first target entry vehicle that matches the target vehicle, including:
[0073] In step S131, a first similarity between the vehicle re-identification feature vector of the target vehicle and the vehicle re-identification feature vectors of each unmatched entry vehicle is calculated;
[0074] In step S132, a second similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle is calculated;
[0075] In step S133, an unmatched entry vehicle whose first similarity is greater than a first preset threshold and whose second similarity is greater than a second preset threshold is determined as a first target entry vehicle.
[0076] Optionally, the first similarity between the vehicle re-identification feature vector of the target vehicle and the re-identification feature vector of each unmatched entry vehicle, and the second similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle can be calculated by using algorithms such as cosine similarity algorithm and Euclidean distance algorithm. It can be understood that the vehicle information may include multiple pieces, so the second similarity between each piece of vehicle information, i.e., vehicle model, vehicle color, and vehicle style, can be calculated separately.
[0077] There is no particular order for step S141 and step S142. Step S141 may be performed first, and then step S142, or step S142 may be performed first, and then step S141, or step S141 and step S142 may be performed simultaneously.
[0078] It can be understood that the unmatched entry vehicles can be sorted in order of entry time from front to back, and the target vehicle can be calculated for similarity with the unmatched entry vehicles in turn according to the sorting results, or an entry time selection interface can be provided to the user corresponding to the target vehicle, and the user can select an entry time range. Based on the entry time range, the unmatched entry vehicles that enter within the entry time range are selected from all the unmatched entry vehicles to perform similarity calculation, thereby reducing the calculation process, improving the calculation speed, and avoiding exit congestion.
[0079] Among them, each unsuccessfully matched entry vehicle refers to an entry vehicle that has not yet been matched for exit or has not been successfully matched.
[0080] Optionally, when a first similarity between the vehicle re-identification feature vector of the target vehicle and the re-identification feature vectors of each unmatched entering vehicle is greater than a first preset threshold, such as 80%, and a second similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entering vehicle is greater than a second preset threshold, such as 85%, then the target vehicle can be considered to be successfully matched with each unmatched entering vehicle, and it is used as the first target entering vehicle. The entry record of the first target entering vehicle is obtained as the entry record of the target vehicle, and the parking fee is calculated for the target vehicle. When the parking fee is paid, the target vehicle can be released.
[0081] In one embodiment of the present application, after the vehicle re-identification feature vector and vehicle information of the target vehicle are matched with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, the following steps are performed:
[0082] When the first target entry vehicle matching the target vehicle is not obtained, weighted average values are calculated for the first similarity and the second similarity to obtain a first overall similarity;
[0083] The entry records of each unmatched entry vehicle are sorted in descending order according to the first overall similarity to obtain a first recommendation list;
[0084] The first recommendation list is recommended to parking lot management personnel.
[0085] It can be understood that the vehicle information may include multiple information, such as vehicle type, vehicle color, and vehicle style, so the second similarity may include multiple information.
[0086] It should be noted that failure to obtain the first target entry vehicle that matches the target vehicle means that the first similarity is less than the first preset threshold, and the second similarity is less than the second preset threshold; or, the first similarity is greater than the first preset threshold, and the second similarity is less than the second preset threshold; or, when the first similarity is less than the first preset threshold, and the second similarity is greater than the second preset threshold, the first target entry vehicle that matches the target vehicle cannot be obtained.
[0087] Optionally, when any similarity does not meet the preset threshold, for example, the first similarity is less than or equal to the first preset threshold, and the second similarity is greater than the second preset threshold, or the first similarity is greater than the first preset threshold, and the second similarity is less than or equal to the second threshold, or the first similarity is less than or equal to the first preset threshold, and the second similarity is less than or equal to the second preset threshold, then the first similarity and the second similarity can be assigned corresponding weights respectively, and the average value is calculated to obtain the overall similarity, that is, the weighted average of the similarities of the re-identified vehicle features, vehicle models, vehicle colors, etc., and then sorted in descending order to obtain a first recommendation list, and the first recommendation list is pushed to the parking lot manager, for example, it can be recommended to customer service so that the customer service can quickly determine the entry record matching the target vehicle, reduce customer service processing costs, and increase release speed.
[0088] It is understandable that when calculating the overall similarity, the license plate similarity may be calculated first and a weight may be assigned to the license plate similarity. However, when the target vehicle is an unlicensed vehicle, the weight assigned to the license plate similarity is 0.
[0089] Among them, if there is no first recommendation list, or there is no record of recommended entry vehicles in the first recommendation list, the customer service needs to notify the owner of the target vehicle to reverse or the owner of the target vehicle needs to report the license plate number before the gate can be charged.
[0090] See also Figure 3 In one embodiment of the present application, before matching the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, the process includes:
[0091] In step S121, it is determined whether the target vehicle is a licensed vehicle;
[0092] In step S122, if yes, it is determined whether the target vehicle is a vehicle with an entry record;
[0093] In step S123, the entry record of the target vehicle is obtained, and based on the entry record of the target vehicle, the target vehicle is released.
[0094] Optionally, the target vehicle can be subjected to license plate recognition. When the license plate is detected, it can be considered as a vehicle with a license plate. At this time, the entry record list can be queried based on its license plate number. If the target vehicle is recorded in the entry record list, it can be determined that the target vehicle is a vehicle with an entry record. At this time, the entry record of the target vehicle can be obtained, and the parking fee can be calculated based on the entry record of the target vehicle. When the parking fee is paid, the target vehicle can be released.
[0095] It is understandable that if the vehicle is an unlicensed vehicle, a temporary license plate or temporary ID may be assigned to the vehicle to independently and uniquely identify the vehicle's identification code for subsequent parking fee calculation and use when the vehicle is released.
[0096] Furthermore, when the target vehicle is a licensed vehicle with an entry record, step S150 can be executed, that is, the parking fee of the target vehicle is charged, and after the parking fee is paid, the gate is opened and the cloud record is uploaded.
[0097] See also Figure 3 In one embodiment of the present application, after determining whether the target vehicle is a vehicle with an entry record, the following steps are included:
[0098] In step S210, if not, the target license plate, vehicle re-identification feature vector and vehicle information of the target vehicle are matched with the license plates, vehicle re-identification feature vectors and vehicle information of each unmatched entry vehicle to determine a second target entry vehicle that matches the target vehicle;
[0099] In step S220, the entry record corresponding to the second target entering vehicle is obtained, and based on the entry record corresponding to the second target entering vehicle, the target vehicle is released.
[0100] Optionally, when the target vehicle is a licensed vehicle, but the entry record of the target vehicle cannot be queried due to problems such as a malfunction of the entry vehicle identification device, the license plate being blocked or the record being invalid when it enters the parking lot, that is, the target vehicle is a vehicle with no entry record. At this time, the parking fee of the target vehicle cannot be calculated. Therefore, the target license plate number of the identified target vehicle, the vehicle re-identification feature vector extracted by the re-identification model, and the vehicle information can be matched with each unmatched entry vehicle, that is, the license plate number, vehicle re-identification feature vector and vehicle information of the unmatched entry vehicles. If all of them are matched successfully, the entry record of the second target entry vehicle that is matched successfully is obtained, the target vehicle is charged based on the entry record, and the gate is opened to release it.
[0101] See also Figure 4In one embodiment of the present application, the target license plate, vehicle re-identification feature vector and vehicle information of the target vehicle are matched with the license plate, vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle to determine a second target entry vehicle that matches the target vehicle, including:
[0102] In step S211, a third similarity between the target license plate of the target vehicle and the license plates of each unmatched entry vehicle is calculated;
[0103] In step S212, a fourth similarity between the vehicle re-identification feature vector of the target vehicle and the vehicle re-identification feature vectors of each unsuccessfully matched entry vehicle is calculated;
[0104] In step S213, a fifth similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle is calculated;
[0105] In step S214, an unmatched entry vehicle whose third similarity is greater than the third preset threshold, whose fourth similarity is greater than the fourth preset threshold, and whose fifth similarity is greater than the fifth preset threshold is determined as a second target entry vehicle.
[0106] Optionally, the third similarity between the target license plate number of the target vehicle and the license plate number of each unmatched entry vehicle, the fourth similarity between the vehicle re-identification feature vector of the target vehicle and the re-identification feature vector of each unmatched entry vehicle, and the fifth similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle can be calculated respectively. It can be understood that the vehicle information can include multiple pieces, so the fourth similarity between each piece of vehicle information, i.e., vehicle model, vehicle color, and vehicle style, can be calculated respectively.
[0107] Among them, the third similarity can be calculated in the following manner. For example, the license plate detection frame image of the target license plate number of the target vehicle can be extracted through the collected video data, and the image is denoised and corrected, and then character segmentation is performed. Then, each segmented character is matched with the standard character set to identify each character, and then sorted in order to obtain the license plate number of the target vehicle. Similarly, the license plates of the unmatched entering vehicles can also be extracted by the above processing, and then the target license plate number is compared with the license plate numbers of the unmatched entering vehicles in turn to determine whether they are consistent.
[0108] The fourth similarity and the fifth similarity may be calculated by using a cosine similarity algorithm, a Euclidean distance algorithm, or other algorithms.
[0109] It can be understood that the unmatched entry vehicles can be sorted in order of entry time from front to back, and the target vehicle can be calculated for similarity with the unmatched entry vehicles in turn according to the sorting results, or an entry time selection interface can be provided to the user corresponding to the target vehicle, and the user can select an entry time range. Based on the entry time range, the unmatched entry vehicles that enter within the entry time range are selected from all the unmatched entry vehicles to perform similarity calculation, thereby reducing the calculation process, improving the calculation speed, and avoiding exit congestion.
[0110] Among them, each unsuccessfully matched entry vehicle refers to an entry vehicle that has not yet been matched for exit or has not been successfully matched.
[0111] Among them, the order of step S211, step S212 and step S213 is not important. Step S211 can be executed first, and then steps S212 and 213, or step S212 can be executed first, and then steps S211 and S213, etc., or steps S211, step S212 and step S213 can be executed simultaneously.
[0112] Optionally, when the third similarity is greater than a third preset threshold, such as 80%, the fourth similarity is greater than a fourth preset threshold, such as 85%, and the fifth similarity is greater than a fifth preset threshold, such as 90%, then the successfully matched and unsuccessfully matched entry vehicles can be obtained as the second target entry vehicle, and the entry record of the second target entry vehicle can be obtained as the entry record of the target vehicle to calculate the parking fee for the target vehicle. When the parking fee is paid, the target vehicle can be released.
[0113] See also Figure 5 In one embodiment of the present application, after matching the target license plate, vehicle re-identification feature vector and vehicle information of the target vehicle with the license plate, vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, the following steps are performed:
[0114] In step S230, when the second target entry vehicle is not obtained, the third similarity, the fourth similarity and the fifth similarity are weighted and averaged to obtain a second overall similarity;
[0115] In step S240, the entry records of each unmatched entry vehicle are sorted in descending order according to the second overall similarity to obtain a second recommendation list;
[0116] In step S250, the second recommendation list is recommended to the parking lot manager.
[0117] It should be noted that failure to obtain the second target entry vehicle that matches the target vehicle refers to a situation where any one or all of the third similarity, the fourth similarity and the fifth similarity do not meet the above-mentioned conditions for matching the second similarity vehicle, for example, the third similarity is less than the third preset threshold, the fourth similarity is less than the fourth preset threshold, and the fifth similarity is less than the fifth preset threshold; or the third similarity is greater than the third preset threshold, the fourth similarity is less than the fourth preset threshold, and the fifth preset threshold is less than the fourth preset threshold; or the third similarity is less than the third preset threshold, and the fourth similarity is greater than the fourth preset threshold, etc., and the first target entry vehicle that matches the target vehicle cannot be obtained.
[0118] Optionally, when any similarity does not meet the preset threshold, for example, the third similarity is less than or equal to the third preset threshold, the fourth similarity is greater than the fourth preset threshold and the fifth similarity is greater than the fifth preset threshold, or the third similarity is greater than the third preset threshold, the fourth similarity is greater than the fourth preset threshold and the fifth similarity is less than or equal to the fifth threshold, or the third similarity is less than or equal to the third preset threshold, the fourth similarity is less than or equal to the fourth preset threshold and the fifth similarity is greater than the fifth preset threshold, etc., then the third similarity, the fourth similarity and the fifth similarity can be respectively assigned corresponding weights, and the average value is calculated to obtain the overall similarity, that is, the weighted average value of the similarities such as the license plate, the re-identified vehicle features, the vehicle model, the vehicle color, etc., and then sorted in descending order to obtain a second recommendation list, and the second recommendation list is pushed to the parking lot manager, for example, it can be recommended to the customer service so that the customer service can quickly determine the entry record matching the target vehicle, reduce the customer service processing cost, and increase the release speed.
[0119] It is understandable that by calculating the similarity of the license plate and assigning a certain weight to the similarity of the license plate, it is possible to ensure that the entry records of vehicles with similar license plates are recommended first. If there is no second recommendation list, or there is no recommended entry vehicle record in the second recommendation list, the customer service needs to notify the owner of the target vehicle to reverse or the owner of the target vehicle needs to report the license plate number before the gate can be charged.
[0120] In one embodiment of the present application, before obtaining the vehicle re-identification feature vector and vehicle information of the target vehicle, the method further includes:
[0121] Determine whether the target vehicle is about to exit;
[0122] If not, determine whether the target vehicle is to enter the venue;
[0123] If so, authenticate the target vehicle;
[0124] When the authentication is successful, the target vehicle is released, and the vehicle information and vehicle re-identification feature vector of the target vehicle are extracted and stored in the entry record.
[0125] Optionally, for entrance and exit channels with the same entry and exit, or in scenarios where entrance and exit barriers are set adjacent to each other, when a target vehicle is identified, it is possible to first determine whether the target vehicle intends to exit or enter, and then perform different operations according to different intentions. For example, when the target vehicle wants to enter, the license plate number of the target vehicle can be identified for authentication. If it is an unlicensed vehicle, a corresponding ID is assigned, and after successful authentication, the target vehicle is released. At the same time, video data is collected through a pre-set vehicle identification device, and the vehicle information and vehicle re-identification feature vector of the target vehicle are extracted based on the video data, and then stored in the entry record for subsequent billing and use when exiting.
[0126] In an embodiment of the present application, when a vehicle enters a parking lot, vehicle information is identified and a vehicle re-identification result is generated and stored. When a vehicle exits a parking lot, a vehicle without a license plate and without an entry record can be matched with the vehicle re-identification result and vehicle information of an unmatched entry vehicle to obtain an entry record with a similarity greater than a threshold, and the fee for the vehicle is calculated and the exit is controlled, thereby achieving rapid release of vehicles without license plates and / or entry records, which can improve user experience, improve the accuracy of parking lot management and control, and reduce parking lot revenue losses.
[0127] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.
[0128] In one embodiment, a vehicle control system is provided, and the vehicle control system corresponds to the vehicle control method in the above embodiment. Figure 6 As shown, the vehicle control system includes an entry vehicle data extraction unit 10, a target vehicle data extraction unit 20, a matching unit 30 and an exit release unit 40. Each functional module is described in detail as follows:
[0129] The vehicle entry data extraction unit 10 is used to collect video data of the area to be identified, and extract vehicle information and vehicle re-identification feature vectors of all vehicles entering the venue in the video data;
[0130] The target vehicle data extraction unit 20 is used to obtain the vehicle re-identification feature vector and vehicle information of the target vehicle when the target vehicle exits the scene;
[0131] The matching unit 30 is used to match the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, so as to obtain a first target entry vehicle that matches the target vehicle;
[0132] The exit release unit 40 is used to obtain the entry record corresponding to the first target entry vehicle, and release the target vehicle based on the entry record of the first target entry vehicle.
[0133] In one embodiment of the present application, the matching unit 40 is further configured to:
[0134] Calculating a first similarity between a vehicle re-identification feature vector of the target vehicle and a vehicle re-identification feature vector of each unsuccessfully matched entry vehicle;
[0135] Calculating a second similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle;
[0136] An unsuccessfully matched entry vehicle whose first similarity is greater than a first preset threshold and whose second similarity is greater than a second preset threshold is determined as a first target entry vehicle.
[0137] In one embodiment of the present application, the system further includes a first recommendation unit, which is used to:
[0138] When the first target entry vehicle matching the target vehicle is not obtained, weighted average values are calculated for the first similarity and the second similarity to obtain a first overall similarity;
[0139] The entry records of each unmatched entry vehicle are sorted in descending order according to the first overall similarity to obtain a first recommendation list;
[0140] The first recommendation list is recommended to parking lot management personnel.
[0141] In one embodiment of the present application, the system further includes a normal exit vehicle determination unit, further configured to:
[0142] Determine whether the target vehicle is a licensed vehicle;
[0143] If so, determine whether the target vehicle is a vehicle with an entry record;
[0144] If so, the target vehicle meets the preset normal exit conditions.
[0145] In one embodiment of the present application, the system further includes a licensed vehicle release unit, which is further used to:
[0146] If the target vehicle has a license plate but no entry record, the target license plate, vehicle re-identification feature vector and vehicle information of the target vehicle are matched with the license plates, vehicle re-identification feature vectors and vehicle information of each unmatched entry vehicle to obtain a second target entry vehicle that matches the target vehicle;
[0147] An entry record corresponding to the second target entering vehicle is obtained, and based on the entry record corresponding to the second target entering vehicle, the target vehicle is released.
[0148] In one embodiment of the present application, there is a license plate vehicle release unit, which is also used for:
[0149] Calculating a third similarity between the target license plate of the target vehicle and the license plates of each unmatched incoming vehicle;
[0150] Calculating a fourth similarity between the vehicle re-identification feature vector of the target vehicle and the vehicle re-identification feature vectors of each unsuccessfully matched entry vehicle;
[0151] Calculating a fifth similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle;
[0152] An unsuccessfully matched entry vehicle whose third similarity is greater than a third preset threshold, whose fourth similarity is greater than a fourth preset threshold, and whose fifth similarity is greater than a fifth preset threshold is determined as a second target entry vehicle.
[0153] In one embodiment of the present application, the system further includes a second recommendation unit, further configured to:
[0154] When the second target entry vehicle is not obtained, weighted averages are respectively performed on the third similarity, the fourth similarity, and the fifth similarity to obtain a second overall similarity;
[0155] The entry records of each unmatched entry vehicle are sorted in descending order according to the second overall similarity to obtain a second recommendation list;
[0156] The second recommendation list is recommended to parking lot management personnel.
[0157] In an embodiment of the present application, when a vehicle enters a parking lot, vehicle information is identified and a vehicle re-identification result is generated and stored. When a vehicle exits a parking lot, a vehicle without a license plate and without an entry record can be matched with the vehicle re-identification result and vehicle information of an unmatched entry vehicle to obtain an entry record with a similarity greater than a threshold, thereby realizing fee calculation and exit control for the vehicle, thereby realizing rapid release of vehicles without license plates and / or entry records, which can improve user experience, improve the accuracy of parking lot management and control, and reduce parking lot revenue losses.
[0158] For the specific definition of the vehicle control system, please refer to the definition of the vehicle control method above, which will not be repeated here. Each module in the above-mentioned vehicle control system can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0159] In one embodiment, a computer device is provided. The computer device may be a terminal device, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium. The readable storage medium stores computer-readable instructions. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer-readable instructions are executed by the processor, a vehicle control method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.
[0160] In an embodiment of the present application, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the steps of the vehicle control method described above are implemented.
[0161] In an embodiment of the application, a readable storage medium is provided, which stores computer-readable instructions. When the computer-readable instructions are executed by a processor, the steps of the vehicle control method described above are implemented.
[0162] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through computer-readable instructions, and the computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they may include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0163] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0164] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.
Claims
1. A vehicle control method, characterized in that: The method comprises: Obtain vehicle information and vehicle re-identification feature vectors of all vehicles entering the area to be identified in the video data; When the target vehicle exits, obtaining a vehicle re-identification feature vector and vehicle information of the target vehicle; Respectively matching the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle to obtain a first target entry vehicle that matches the target vehicle; An entry record corresponding to the first target entry vehicle is obtained, and based on the entry record corresponding to the first target entry vehicle, the target vehicle is released.
2. The vehicle control method according to claim 1, characterized in that: The step of matching the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle to obtain a first target entry vehicle that matches the target vehicle includes: Calculating a first similarity between the vehicle re-identification feature vector of the target vehicle and the vehicle re-identification feature vectors of each unsuccessfully matched entry vehicle; Calculating a second similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle; An unsuccessfully matched entry vehicle whose first similarity is greater than a first preset threshold and whose second similarity is greater than a second preset threshold is determined as the first target entry vehicle.
3. The vehicle control method according to claim 2, characterized in that: After the vehicle re-identification feature vector and vehicle information of the target vehicle are matched with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, the method further comprises: When a first target entry vehicle matching the target vehicle is not obtained, weighted average values are taken for the first similarity and the second similarity to obtain a first overall similarity; Sorting the admission records of the unmatched admission vehicles in descending order of the first overall similarity to obtain a first recommendation list; The first recommendation list is recommended to parking lot management personnel.
4. The vehicle control method according to claim 1, characterized in that: Before respectively matching the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, the method includes: Determining whether the target vehicle is a licensed vehicle; If yes, determining whether the target vehicle is a vehicle with an entry record; If yes, the entry record of the target vehicle is obtained, and based on the entry record of the target vehicle, the target vehicle is released.
5. The vehicle control method according to claim 4, characterized in that: After determining whether the target vehicle is a vehicle with an entry record, the method includes: If not, the target license plate, the vehicle re-identification feature vector and the vehicle information of the target vehicle are matched with the license plates, the vehicle re-identification feature vectors and the vehicle information of each of the unmatched entry vehicles to obtain a second target entry vehicle that matches the target vehicle; An entry record corresponding to the second target entering vehicle is obtained, and based on the entry record corresponding to the second target entering vehicle, the target vehicle is released.
6. The vehicle control method according to claim 5, characterized in that: The step of matching the target license plate, the vehicle re-identification feature vector, and the vehicle information of the target vehicle with the license plate, the vehicle re-identification feature vector, and the vehicle information of each of the unmatched entry vehicles to obtain a second target entry vehicle that matches the target vehicle includes: Calculating a third similarity between the target license plate of the target vehicle and the license plates of each of the unmatched entry vehicles; Calculating a fourth similarity between the vehicle re-identification feature vector of the target vehicle and the vehicle re-identification feature vectors of each of the unsuccessfully matched entry vehicles; Calculating a fifth similarity between the vehicle information of the target vehicle and the vehicle information of each unmatched entry vehicle; An unsuccessfully matched entry vehicle whose third similarity is greater than a third preset threshold, whose fourth similarity is greater than a fourth preset threshold, and whose fifth similarity is greater than a fifth preset threshold is determined as the second target entry vehicle.
7. The vehicle control method according to claim 6, characterized in that: After matching the target license plate, the vehicle re-identification feature vector and the vehicle information of the target vehicle with the license plate, the vehicle re-identification feature vector and the vehicle information of each unmatched entry vehicle, the method further comprises: When the second target entering vehicle is not obtained, weighted average values are respectively obtained for the third similarity, the fourth similarity and the fifth similarity to obtain a second overall similarity; Sorting the admission records of the unmatched admission vehicles in descending order of the second overall similarity to obtain a second recommendation list; The second recommendation list is recommended to parking lot management personnel.
8. A vehicle control system, characterized in that: The system comprises: An entry vehicle data extraction unit, used to obtain vehicle information and vehicle re-identification feature vectors of all entry vehicles in the video data of the area to be identified; A target vehicle data extraction unit, used for obtaining a vehicle re-identification feature vector and vehicle information of the target vehicle when the target vehicle exits the scene; A matching unit, used to match the vehicle re-identification feature vector and vehicle information of the target vehicle with the vehicle re-identification feature vector and vehicle information of each unmatched entry vehicle, so as to obtain a first target entry vehicle that matches the target vehicle; The exit release unit is used to obtain the entry record corresponding to the first target entry vehicle, and release the target vehicle based on the entry record corresponding to the first target entry vehicle.
9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the vehicle control method steps as described in any one of claims 1 to 7 are implemented.
10. A readable storage medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by a processor, the steps of the vehicle control method according to any one of claims 1 to 7 are implemented.
Citation Information
Patent Citations
Unlicensed vehicle charging method and device
CN107730617A
Parking management method and device, computer equipment and storage medium
CN112396708A
Vehicle identification method, device and system, and storage medium
CN114663862A
Multi-node collaborative unlicensed vehicle parking management method
CN114764951A
Parking lot unlicensed vehicle passing management method and system, electronic equipment and storage medium
CN115861974A
Cited By
Intelligent identification and release method and system based on vehicle type
CN121617256A
Intelligent identification and release method and system based on vehicle type
CN121617256B
