An algorithm-assisted field library bit number correction method

Abstract

The present application relates to the field of data processing, and particularly relates to a kind of field library residual bit number correction method based on algorithm assistance, specifically includes: data acquisition step, data processing step, counting step, server passes through license plate number confirmation current reported passage record whether with the passage type of the passage record of previous report exists conflict, if there is conflict, carry out first verification step;For the exit record of unable to match entry record, carry out second verification step.In the absence of the participation of other field library information, based on the vehicle passage record in the collected image, avoid the error correction of field library residual bit number based on unreliable, incomplete vehicle entry and exit record data.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology, specifically relating to an algorithm-assisted field library remainder correction method. Background Technology

[0002] After the parking lot went online, oversights in the parking lot management system or violations by vehicle owners caused discrepancies in the number of available parking spaces, making it difficult to guarantee long-term accuracy. Existing methods for correcting these discrepancies primarily rely on the completeness of vehicle entry and exit records. However, in practice, the accuracy and reasonableness of some records cannot be guaranteed, and correcting discrepancies based on these records may actually lead to even greater deviations in the parking lot's available parking spaces. Summary of the Invention

[0003] To address the above technical problems, this invention proposes an algorithm-assisted field library remainder correction method, including data acquisition steps, data processing steps, and counting steps:

[0004] In the data acquisition step, the acquisition module of the parking garage terminal device captures image data of each vehicle entering and exiting the parking garage and sends the image data to the processing module of the parking garage terminal device.

[0005] The data processing steps include the following steps implemented by the processing module of the warehouse terminal equipment:

[0006] The license plate recognition algorithm is invoked to identify the license plate numbers of passing vehicles from the image data, and a passage record including the identification number is generated;

[0007] The passage record is associated with the image data and then reported to the server; the passage record includes the license plate number information of the identified passing vehicle;

[0008] The system uses algorithms including moving object detection and tracking to check whether the vehicle's entry and exit behavior is complete. Once the completeness of the vehicle's entry and exit behavior is confirmed, a vehicle-to-motor vehicle detection algorithm is used to determine whether the vehicle is a motor vehicle from the image data.

[0009] The image data processing results are added to the passage record and then asynchronously reported to the server; the processing results include the entry and exit records of the passing vehicles;

[0010] The counting step includes the following steps implemented by the server:

[0011] Based on the processing results, the number of remaining warehouse occupancy numbers can be increased, decreased, or corrected.

[0012] Send the updated field inventory balance to the field inventory management terminal.

[0013] Without requiring the participation of other parking lot information, the system relies on vehicle passage records in the collected images to avoid incorrect correction of parking lot digits based on unreliable and incomplete vehicle entry and exit records. Attached Figure Description

[0014] Figure 1 : A schematic diagram of the equipment distribution in one embodiment;

[0015] Figure 2 A flowchart of one implementation method;

[0016] Figure 3 : Flowchart of a method for correcting the remaining position of the field storage in one embodiment. Detailed Implementation

[0017] The following embodiments further illustrate the content of the present invention, but should not be construed as limiting the present invention. Any modifications or substitutions made to the methods, steps, or conditions of the present invention without departing from the spirit and essence of the invention are within the scope of the present invention.

[0018] The term "garage terminal equipment" refers to electronic network equipment installed at the entrances and exits of parking lots and garages, which has the ability to capture image data and process certain data, such as common license plate recognition cameras.

[0019] Some implementation methods are as follows Figure 1 The layout includes a facility terminal device (containing a data acquisition module and a processing module) installed inside the facility, a server communicating with it, and a facility management terminal, to complete tasks such as... Figure 2 The steps shown include: data acquisition, data processing, and counting.

[0020] In the data acquisition step, the acquisition module of the parking garage terminal device captures image data of each vehicle entering and exiting the parking garage and sends the image data to the processing module of the parking garage terminal device.

[0021] The data processing steps include the following steps implemented by the processing module of the warehouse terminal equipment:

[0022] The license plate recognition algorithm is invoked to identify the license plate numbers of passing vehicles from the image data, and a passage record including the identification number is generated;

[0023] The passage record is associated with the image data and then reported to the server; the passage record includes the license plate number information of the identified passing vehicle;

[0024] The system uses algorithms including moving object detection and tracking to check whether the vehicle's entry and exit behavior is complete. Once the completeness of the vehicle's entry and exit behavior is confirmed, a vehicle-to-motor vehicle detection algorithm is used to determine whether the vehicle is a motor vehicle from the image data.

[0025] The image data processing results are added to the passage record and then asynchronously reported to the server; the processing results include the entry and exit records of the passing vehicles;

[0026] The counting step includes the following steps implemented by the server:

[0027] Based on the processing results, the number of remaining warehouse occupancy numbers can be increased, decreased, or corrected.

[0028] Send the updated field inventory balance to the field inventory management terminal.

[0029] The counting steps in a more specific implementation also include the following steps:

[0030] The server uses the license plate number to confirm whether the currently reported passage record conflicts with the entry / exit type of the previously reported passage record. If a conflict exists, the server performs the first verification step.

[0031] The server calls three types of image data associated with the current passage record, the passage record that conflicts with it, and the passage records under the historical license plate number.

[0032] Use enhanced license plate recognition algorithms and / or vehicle accessory recognition algorithms to reconfirm the license plate number or vehicle accessory information in the above three types of image data;

[0033] The current passage record, conflicting passage records, and historical passage records for the same license plate number are compared with each other. Abnormal passage records are marked to provide a basis for subsequent correction.

[0034] In some specific implementations, the counting step also includes a second verification step for exit records that cannot be matched with entry records. That is, the server calculates the similarity between the exiting vehicle and the vehicles present using a vehicle face matching algorithm to provide auxiliary judgment for manual confirmation.

[0035] In some more specific implementations, the second verification step involves the server extracting the vehicle face features of the departing vehicle and the vehicle face features of the present vehicle from the vehicle image data and comparing the two.

[0036] When there is a vehicle in the parking lot with a different license plate but a similarity to the exiting vehicle that exceeds a preset threshold, a notification is sent to the parking lot management terminal.

[0037] The counting steps in some specific implementations include:

[0038] When it is confirmed that the omission is due to a missing vehicle entry and exit record in the parking lot, the remaining number of parking lot digits will be corrected.

[0039] When the processing result indicates a duplicate entry, the number of remaining digits in the storage area will be increased by 1.

[0040] When the processing result indicates a duplicate appearance, the number of remaining digits in the field inventory is reduced by 1.

[0041] Some specific implementation methods also include the following counting steps: modifying the remaining number of the warehouse through the warehouse management terminal. For example, the server notifies the warehouse management terminal administrators online to manually confirm the cause of the anomaly and then decide whether correction is needed.

[0042] The method of the present invention will be explained below through a specific embodiment.

[0043] The method for correcting the remaining number of digits in this embodiment employs, as follows: Figure 3 The process shown is as follows: First, the terminal equipment in the parking lot captures image data (captured images) of vehicles entering and exiting, and uses a license plate recognition algorithm to obtain the license plate number of the passing vehicle, generating a passage record with a unique identifier and containing information such as the license plate number, and attaching image data to report to the server.

[0044] The server adjusts the number of available slots in the storage area based on the entry and exit types of the received access records.

[0045] The terminal equipment in the parking lot captures continuous image data of vehicles entering and exiting the parking lot, and uses moving object detection and tracking algorithms to detect whether the vehicle's entry and exit behavior is complete. It then attaches a passage record identifier to this result and asynchronously reports it to the server.

[0046] When the server receives a report of abnormal vehicle entry and exit behavior, it corrects the remaining digits in the relevant parking lot.

[0047] After the server confirms the integrity of the entry and exit behavior, the image data in the passage record is processed using a non-motorized vehicle detection algorithm to determine whether the passing vehicle is a motorized vehicle. If it is a non-motorized vehicle, the remaining number of digits in the parking lot is corrected.

[0048] The license plate number is used to confirm whether the currently reported passage record conflicts with the previous passage record in terms of entry and exit type. If so, an enhanced license plate recognition algorithm and a vehicle product recognition algorithm are used to process the image data associated with the current passage record, the image data associated with passage records that have logical conflicts with it, and the image data associated with the historical passage records of the same license plate number. This confirms the correctness of the license plate number recognition of the three and obtains the vehicle product information of the three. By comparing the differences among the three, abnormal passage records can be quickly located, providing a basis for subsequent correction.

[0049] For vehicle exit records that cannot be matched with entry records, a vehicle face matching algorithm can be used to obtain the similarity between exiting vehicles and vehicles present, providing auxiliary judgment for subsequent manual confirmation steps.

[0050] All algorithms used in this embodiment can be purchased from commercial cloud computing platforms.

[0051] Based on the processing results of various algorithms on the image data associated with the passage records, the number of remaining bits in the field library can be directly corrected, or a message can be sent online to the management terminal to notify the management personnel to confirm the cause of the anomaly before deciding whether correction is necessary.

[0052] In this embodiment, the terminal device refers to an electronic network device installed at the entrance and exit of the parking lot, which has the ability to capture image data and process certain data, such as a common license plate recognition camera.

[0053] In this embodiment, the unique identifier of the passage record is composed of the parking lot identifier, the lane identifier, and the time of the captured image.

[0054] In this embodiment, a moving object detection and tracking algorithm is used to confirm the integrity of vehicle entry and exit behavior. When the parking management system allows a vehicle to pass but it does not pass or turns back midway, the remaining number of parking spaces is calculated based on the entry and exit types of the passage records, which can lead to a deviation in the number of parking spaces. When a passage record marked by the algorithm for this type of behavior is found, the number of parking spaces will be corrected. If the parking entrance is allowed but the vehicle does not actually enter, the incorrect decrease in the number of parking spaces will be corrected. If the parking exit is allowed but the vehicle does not actually leave, the incorrect increase in the number of parking spaces will be corrected.

[0055] In this embodiment, a vehicle-to-non-vehicle detection algorithm is used to confirm that the main object of passage is a motor vehicle. When the main object of passage is not a motor vehicle and has a counterfeit license plate, the license plate recognition camera in the parking lot system is easily misled into reporting the passage record, causing the parking lot management system to incorrectly increase or decrease the number of digits. When a record marked by the algorithm for this type of behavior is found, the parking lot digits will be corrected to correct the erroneous increase or decrease in the parking lot digits.

[0056] This embodiment uses license plate recognition algorithms and vehicle accessory recognition algorithms to confirm the uniqueness of vehicles in time and space. For example, when a vehicle enters or exits, if the entry / exit type matches the previous historical record of that license plate in the parking lot (i.e., the vehicle has entered or exited repeatedly), it is necessary to verify the correctness of the license plate and vehicle accessories in the current record. Since the hardware resources of terminal devices in the parking lot are generally less than those of servers, when a record type conflict occurs, the server performs an enhanced license plate recognition algorithm to avoid errors in license plate recognition by the terminal devices. If the license plate number recognized by the enhanced license plate recognition algorithm on the server and the terminal equipment in the parking lot is still consistent, then vehicle identification is performed on the vehicle images in the spatiotemporally conflicting entry records, and also on the image data in the historical records. By comparing the vehicle's entry and exit records in the current parking lot, the conflicting entry and exit records, and the historical entry and exit records, it is determined whether the license plate is counterfeited or damaged, resulting in the same license plate number, or whether the vehicle's entry and exit records in the parking lot were missed. Only when it is confirmed that the vehicle's entry and exit records in the parking lot were missed, the parking lot digit remainder was corrected. If it is determined to be a duplicate entry, the digit remainder is increased by 1; if it is determined to be a duplicate exit, the digit remainder is decreased by 1.

[0057] In this embodiment, a vehicle face matching algorithm is used to assist in maintaining the integrity of vehicle entry and exit records. When a vehicle leaves the parking lot, the vehicle face features are extracted from the currently captured vehicle image data and compared with the vehicle face features extracted from the image data of other vehicles in the parking lot. If there is a vehicle in the parking lot with a different license plate but a similarity higher than a preset threshold, the on-duty personnel are notified online. The on-duty personnel then confirm whether the vehicle has violated regulations by changing its license plate number, thereby ensuring the integrity of the entry and exit records and avoiding incorrect correction of parking lot space.

[0058] The embodiments and functional operations of the subject matter described in this specification can be implemented in the following ways: digital electronic circuits, tangibly implemented computer software or firmware, computer hardware, including the structures disclosed in this specification and their equivalents, or combinations thereof. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on one or more tangible, non-transitory program carriers, for execution by a data processing device or to control the operation of the data processing device. Alternatively or additionally, the program instructions can be encoded on artificially generated propagation signals, such as machine-generated electrical signals, optical signals, or electromagnetic signals, which are generated as encoded information to be transmitted to an appropriate receiver device executed by the data processing device. The computer storage medium can be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or one or more combinations thereof.

[0059] Specific implementations of the subject matter have been described. Other implementations are within the scope of the following claims. For example, the activities described in the claims can be performed in a different order and still achieve the desired result. As an example, the processes described in the drawings do not necessarily require a specific order or sequence to be shown in order to achieve the desired result. In certain implementations, multitasking and parallel processing may be advantageous.

[0060] Although the present invention has been described in detail above with general descriptions, specific embodiments, and experiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A method for correcting the number of remaining positions in a field bank based on an algorithm, characterized in that, This includes data acquisition steps, data processing steps, and counting steps: In the data acquisition step, the acquisition module of the parking garage terminal device captures image data of each vehicle entering and exiting the parking garage and sends the image data to the processing module of the parking garage terminal device. The data processing steps include the following steps implemented by the processing module of the warehouse terminal equipment: The license plate recognition algorithm is invoked to identify the license plate numbers of passing vehicles from the image data, and a passage record including the identification number is generated; The passage record is associated with the image data and then reported to the server; the passage record includes the license plate number information of the identified passing vehicle; The system uses algorithms including moving object detection and tracking to check whether the vehicle's entry and exit behavior is complete. Once the completeness of the vehicle's entry and exit behavior is confirmed, a vehicle-to-motor vehicle detection algorithm is used to determine whether the vehicle is a motor vehicle from the image data. The image data processing results are added to the passage record and then asynchronously reported to the server; the processing results include the entry and exit records of the passing vehicles; The counting step includes the following steps implemented by the server: Based on the processing results, the remaining number of digits in the warehouse is corrected, including increases and decreases. Send the updated field inventory balance to the field inventory management terminal.

2. The method as described in claim 1, characterized in that, The counting step also includes the following steps: The server uses the license plate number to confirm whether the currently reported passage record conflicts with the entry / exit type of the previously reported passage record. If a conflict exists, the server performs the first verification step. For exit records that cannot be matched with entry records, a second verification step is performed.

3. The method as described in claim 2, characterized in that, The first verification step includes the following steps: The server calls three types of image data associated with the current passage record, the passage record that conflicts with it, and the passage records under the historical license plate number. The license plate number in the three types of image data was reconfirmed using an enhanced license plate recognition algorithm, and the vehicle product information in the three types of image data was reconfirmed using a vehicle product recognition algorithm. The current passage record, conflicting passage records, and historical passage records for the same license plate number are compared with each other, and abnormal passage records are marked to provide a basis for correction in the counting step.

4. The method as described in claim 2, characterized in that, The second verification step includes the following steps: The server uses a vehicle face matching algorithm to calculate the similarity between vehicles leaving the vehicle and vehicles present, in order to provide auxiliary judgment for manual confirmation.

5. The method as described in claim 4, characterized in that, The second verification step specifically includes the following steps: Extract the vehicle face features of the departing vehicle and the vehicle face features of the present vehicle from the vehicle image data, and compare the two. When there is a vehicle in the parking lot with a different license plate but a similarity to the exiting vehicle that exceeds a preset threshold, a notification is sent to the parking lot management terminal.

6. The method as described in claim 1, characterized in that, The identification number is generated by combining the parking lot identification number, lane identification number, and the time of the captured image.

7. The method as described in claim 1, characterized in that, The image data includes captured images and continuous images.

8. The method as described in claim 1, characterized in that, The counting step specifically includes: When it is confirmed that the omission is due to a missing vehicle entry and exit record in the parking lot, the remaining number of parking lot digits will be corrected. When the processing result indicates a duplicate entry, the number of remaining digits in the storage area will be increased by 1. When the processing result indicates a duplicate appearance, the number of remaining digits in the field inventory is reduced by 1.

9. The method as described in claim 1, characterized in that, The counting step also includes: modifying the remaining number of the warehouse through the warehouse management terminal.

10. A parking lot digit correction system, comprising a parking lot terminal device including a data acquisition module and a processing module installed inside the parking lot, a server communicating with the terminal device, and a parking lot management terminal, characterized in that, The warehouse terminal device, the server, and the warehouse management terminal together implement the steps in any one of the methods as described in claims 1-9.

Images