Method for calculating recognition accuracy of ETC portal snapshot camera

Abstract

The invention discloses a method for calculating the recognition accuracy of an ETC portal frame snapshot camera, and relates to the technical field of expressways, and the method comprises the following steps: firstly, restoring a vehicle driving path through recognizing a license plate and a toll station license plate; deducing a theoretical license plate based on the driving path; finally, the recognized license plate and the theoretical license plate are compared and calculated, the recognition accuracy is obtained, through the method, under the condition of pure ETC door frame license plate recognition recording and toll station recording, the snapshot accuracy of the door frame camera can be judged, and in the process, personnel do not need to conduct secondary verification on pictures captured by the camera. The method is also convenient for programming realization based on a software technology, the accuracy of the camera is monitored in an automatic manner, the cost of human resources and secondary identification of the equipment is effectively reduced, a high-speed owner unit can conveniently repair and process the equipment with unqualified accuracy in time, and the working efficiency of the high-speed owner unit is improved. And a camera manufacturer can conveniently carry out targeted optimization on the identification accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of expressways, and more specifically, to a method for calculating the recognition accuracy of an ETC gantry snapshot camera. Background Art

[0002] In 2019, the national highway toll collection mode was changed to free-flow toll collection, and the ETC gantry played a great role in highway toll transactions. The ETC gantry is equipped with an RSU antenna and a snapshot camera, which are used for gantry transactions and vehicle toll snapshots respectively. The snapshot camera mainly has the functions of capturing vehicle toll images and license plate recognition. The low data accuracy not only affects the splitting and accounting of highway tolls, but also interferes with the highway operating unit's correct judgment of road traffic conditions.

[0003] The function status of the snapshot camera mainly depends on the device hardware and parameter settings, but it will ultimately be reflected in the recognition accuracy. The recognition accuracy will also continue to decline as the device ages.

[0004] Conventional method for calculating recognition accuracy:

[0005] Method 1: Manual on-site license plate counting: Manual on-site (or relying on on-site video recording) to count the license plates, record the license plate numbers, and then compare and analyze them with the license plates recognized by the camera to calculate the recognition accuracy.

[0006] Method 2: Re-recognize the captured images. When the default capture camera captures all vehicle passing images, the image results are re-recognized and checked manually or with text recognition software to calculate the recognition accuracy of the original capture camera.

[0007] Effect: Manual methods are labor-intensive and difficult to sustain. Conventional text recognition software has a low accuracy rate and is difficult to verify the accuracy of the original capture camera.

[0008] Based on the above problems, we provide a method to calculate the recognition accuracy of the ETC gantry snapshot camera.

[0009] The above information disclosed in this background technology is only used to increase the understanding of the background technology of the present invention and therefore, it may include information that does not constitute the prior art known to ordinary technicians in this field. Summary of the invention

[0010] In order to solve the problems raised in the above background technology, the present invention provides a method for calculating the recognition accuracy of the ETC gantry snapshot camera.

[0011] The present invention provides a method for calculating the recognition accuracy of an ETC gantry snapshot camera using the following technical solution:

[0012] A method for calculating the recognition accuracy of an ETC gantry snapshot camera comprises the following steps:

[0013] S1. First, the vehicle's driving path is restored by identifying the license plate and the toll booth license plate;

[0014] S2, then, derive the theoretical license plate based on the driving path;

[0015] S3. Finally, the recognized license plate is compared with the theoretical license plate to obtain the recognition accuracy.

[0016] Preferably, the vehicle driving path is divided into an out-of-town vehicle path, a single station path, and an entry-exit path;

[0017] The route of non-local vehicles refers to the route that has no record of passing the toll station of this road section;

[0018] A single-station route refers to a route that only has a toll station entrance pass record, or only has a toll station exit pass record;

[0019] The entry and exit path refers to the path that has both the toll station entrance and toll station exit passage records.

[0020] Preferably, in S1, the path of the out-of-town vehicle is calculated. First, the difference between the recognized license plate and the toll station license plate is taken:

[0021] P 外地车车牌 =P 识别车牌 -P 收费站车牌

[0022] Among the out-of-town license plates, the license plates that appear at least twice in the same direction;

[0023] Get all the license plates of non-local vehicles generated in a certain direction of travel;

[0024] The license plate of the out-of-town vehicle is the theoretical license plate of all ETC gantries in that direction.

[0025] Preferably, the single-station path is calculated by first taking the recorded time of the toll station exit as the end time, searching for the latest recorded license plate before the vehicle leaves the highway on two possible paths, and taking the path where the record is located as the vehicle travel path;

[0026] Based on this driving path, find the ETC gantries that the vehicle passes through in sequence;

[0027] Based on the driving path, find the ETC gantries that the vehicle passes through in sequence, and use the license plate as the theoretical license plate of the ETC gantry.

[0028] Preferably, the single-station path is calculated by first taking the toll station entrance record time as the starting time, searching for the earliest recorded license plate after the vehicle enters the highway on two possible paths, and taking the path where the record is located as the vehicle travel path;

[0029] Based on this driving path, find the ETC gantries that the vehicle passes through in sequence;

[0030] Based on the driving path, find the ETC gantries that the vehicle passes through in sequence, and use the license plate as the theoretical license plate of the ETC gantry.

[0031] Preferably, the entry and exit paths are calculated with reference to the calculation of the single-station path; based on the driving path, the ETC gantries that the vehicle passes through in sequence are searched, and the license plate is used as the theoretical license plate of the ETC gantry.

[0032] Preferably, the theoretical license plate of each ETC gantry is obtained based on the out-of-town vehicle path, the single station path, and the entry and exit path, and compared with the recognized license plate of the gantry, and the accuracy is obtained as follows:

[0033] Taking the theoretical license plate in the identified license plate as the numerator and the theoretical license plate as the denominator, calculate the accuracy rate:

[0034]

[0035] Taking the theoretical license plate in the recognized license plate as the numerator and the recognized license plate as the denominator, calculate the actual shooting accuracy:

[0036]

[0037] In summary, the present invention includes the following beneficial technical effects:

[0038] This method can be used to determine the capture accuracy of the gantry camera in the case of pure ETC gantry plate recognition records and toll station records, without the need for personnel to conduct secondary verification of the camera captured images. This method is also easy to implement programming based on software technology, and monitors the camera accuracy in an automated manner, effectively reducing the cost of human resources and equipment secondary recognition, making it easier for highway owners to repair and process equipment that does not meet the accuracy standards in a timely manner, and also easier for camera manufacturers to optimize the recognition accuracy.

[0039] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present invention will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the highway;

[0041] Figure 2 It is a schematic diagram of the route of out-of-town vehicles;

[0042] Figure 3 It is a single station path diagram;

[0043] Figure 4 It is a schematic diagram of the entry and exit routes. DETAILED DESCRIPTION

[0044] The following is combined with Figures 1 to 4 The present invention is described in further detail.

[0045] It should be noted that the drawings are schematic and not to scale. For the sake of clarity and convenience, the relative sizes and proportions of the parts shown in the drawings are exaggerated or reduced in size, and any size is only illustrative and not restrictive. In addition, the same reference symbol is used for the same structure, element or accessory appearing in more than two figures to reflect similar features.

[0046] The embodiment of the present invention discloses a method for calculating the recognition accuracy of an ETC gantry snapshot camera. Figures 1 to 4 A method for calculating the recognition accuracy of an ETC gantry snapshot camera, under the condition that the default snapshot camera captures all vehicle passing pictures, includes the following steps:

[0047] S1. First, the vehicle's driving path is restored by identifying the license plate and the toll booth license plate;

[0048] S2, then, derive the theoretical license plate based on the driving path;

[0049] S3. Finally, the recognized license plate is compared with the theoretical license plate to obtain the recognition accuracy.

[0050] Worth noting:

[0051] One-time pass: A vehicle enters the expressway through the toll station entrance, drives a certain distance and then exits the expressway through the toll station exit, which is counted as one-time pass;

[0052] Path fitting: Although highway operators can obtain the ETC gantry's license plate recognition and transaction data, and toll station traffic data, they cannot directly associate the data with vehicle paths. Associating traffic data with traffic paths based on data analysis is called path fitting;

[0053] License plate recognition: The camera recognizes the license plate;

[0054] Toll station license plate: the license plate in the toll station traffic data;

[0055] Theoretical license plate: The theoretical license plate of the vehicle passing through the camera is derived based on the driving path.

[0056] Reference Figure 1 It has jurisdiction over 5 toll stations, which are denoted by A~E respectively; AE is the upward direction and EA is the downward direction; near the toll station, there is a set of gantries in the upward and downward directions respectively, which are denoted by A1~E1 and A2~E2 respectively.

[0057] Vehicle driving paths are divided into out-of-town vehicle paths, single-station paths, and entry-exit paths;

[0058] The route of non-local vehicles refers to the route that has no record of passing the toll station of this road section;

[0059] A single-station route refers to a route that only has a toll station entrance pass record, or only has a toll station exit pass record;

[0060] The entry and exit path refers to the path with both the toll station entrance and toll station exit pass records;

[0061] The above classification is based on the data within a limited time range.

[0062] Figure 2 For the two routes for out-of-town vehicles, vehicles enter the expressway from the right end (or left end) of the road, and leave the expressway after passing through E1~A1 (or A2~E2) without passing through the toll station of this expressway.

[0063] In S1, the license plate of the out-of-town vehicle is calculated. First, the difference between the recognized license plate and the toll booth license plate is taken:

[0064] P 外地车车牌 =P 识别车牌 -P 收费站车牌

[0065] Among the out-of-town license plates, the license plates that appear at least twice in the same direction (AE direction or EA direction) are selected. (Purpose: to eliminate the redundant license plates generated by the capture camera's misrecognition);

[0066] Get all the license plates of non-local vehicles generated in a certain direction of travel;

[0067] The license plate of the out-of-town vehicle is the theoretical license plate of all ETC gantries in that direction.

[0068] Reference Figure 3 ,There are 4 paths that generate single-station records, 2 from the AE direction and 2 from the EA direction.

[0069] To calculate the single-station path, first use the toll station exit record time as the end time, and search for the latest license plate recognition record before the vehicle leaves the highway on two possible paths, and use the path where this record is located as the vehicle's driving path;

[0070] Based on this driving path, find the ETC gantries that the vehicle passes through in sequence;

[0071] If there are still recognized license plates after eliminating the ETC gantry recognition license plate records (based on this driving path and then searching for the ETC gantry that the vehicle passes through in sequence), it will be processed according to the out-of-town vehicle path;

[0072] Based on the driving path, find the ETC gantries that the vehicle passes through in sequence, and use the license plate as the theoretical license plate of the ETC gantry.

[0073] Specifically, to calculate the single-station path, first take the toll station entrance record time as the starting time, and search for the earliest recorded license plate after the vehicle enters the highway on two possible paths, and use the path where this record is located as the vehicle's driving path;

[0074] Based on this driving path, find the ETC gantries that the vehicle passes through in sequence;

[0075] If there are still recognized license plates after eliminating the ETC gantry recognition license plate records (based on this driving path and then searching for the ETC gantry that the vehicle passes through in sequence), it will be processed according to the out-of-town vehicle path;

[0076] Based on the driving path, find the ETC gantries that the vehicle passes through in sequence, and use the license plate as the theoretical license plate of the ETC gantry.

[0077] Reference Figure 4 All types of toll station entrance and exit records can be ultimately divided into five types of travel paths.

[0078] Except for the first category, the other four categories should all belong to single-station paths, so we only need to judge whether the first category of paths is met.

[0079] by Figure 4 Taking the first type of path as an example, the first type of path must meet the following conditions:

[0080] (1) The entry time of station D is less than the exit time of station B, and there are no entry and exit records of other stations within this time range;

[0081] (2) Taking the entry time of station D as the starting time, search for the earliest license plate recognition record after the vehicle enters the expressway, and the corresponding ETC gantry is located on the path D→D1→C1→B;

[0082] (3) Taking the exit time of station B as the end time, find the latest license plate recognition record before the vehicle leaves the expressway. The corresponding ETC gantry is located on the path D→D1→C1→B.

[0083] Based on the driving path, find the ETC gantries that the vehicle passes through in sequence, and use the license plate as the theoretical license plate of the ETC gantry.

[0084] Specifically, based on the out-of-town vehicle path, single station path, and entry and exit paths, the theoretical license plate of each ETC gantry is obtained, and it is compared with the recognized license plate of the gantry, and the accuracy is as follows:

[0085] Taking the theoretical license plate in the identified license plate as the numerator and the theoretical license plate as the denominator, calculate the accuracy rate:

[0086]

[0087] Taking the theoretical license plate in the recognized license plate as the numerator and the recognized license plate as the denominator, calculate the actual shooting accuracy:

[0088]

[0089] This method can be used to determine the capture accuracy of the gantry camera in the case of pure ETC gantry plate recognition records and toll station records, without the need for personnel to conduct secondary verification of the camera captured images. This method is also easy to program based on software technology and monitor the camera accuracy in an automated manner.

[0090] Through this method, the cost of human resources and equipment secondary identification can be effectively reduced.

[0091] The beneficial effects of judging accuracy: it is convenient for highway owners to promptly repair and process equipment that does not meet accuracy standards, and it is also convenient for camera manufacturers to carry out targeted optimization of recognition accuracy.

[0092] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. "Multiple" means two or more, unless otherwise clearly and specifically defined.

[0093] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0094] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0095] In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other.

[0096] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for calculating the recognition accuracy of an ETC gantry snapshot camera, characterized in that: The following steps are involved: S1. First, the vehicle's driving path is restored by identifying the license plate and the toll booth license plate; S2, then, derive the theoretical license plate based on the driving path; S3. Finally, the recognized license plate is compared with the theoretical license plate to obtain the recognition accuracy.

2. According to the method for calculating the recognition accuracy of the ETC gantry snapshot camera according to claim 1, it is characterized in that: Vehicle driving paths are divided into out-of-town vehicle paths, single-station paths, and entry-exit paths; The route of non-local vehicles refers to the route that has no record of passing the toll station of this road section; A single-station route refers to a route that only has a toll station entrance pass record, or only has a toll station exit pass record; The entry and exit path refers to the path that has both the toll station entrance and toll station exit passage records.

3. According to the method for calculating the recognition accuracy of the ETC gantry snapshot camera according to claim 2, it is characterized in that: In S1, the path of the out-of-town vehicle is calculated. First, the difference between the recognized license plate and the toll station license plate is taken: P 外地车车牌 =P 识别车牌 -P 收费站车牌 Among the out-of-town license plates, the license plates that appear at least twice in the same direction; Get all the license plates of non-local vehicles generated in a certain direction of travel; The license plate of the out-of-town vehicle is the theoretical license plate of all ETC gantries in that direction.

4. According to the method for calculating the recognition accuracy of the ETC gantry snapshot camera according to claim 2, it is characterized in that: To calculate the single-station path, first use the toll station exit record time as the end time, and search for the latest license plate recognition record before the vehicle leaves the highway on two possible paths, and use the path where this record is located as the vehicle's driving path; Based on this driving path, find the ETC gantries that the vehicle passes through in sequence; Based on the driving path, find the ETC gantries that the vehicle passes through in sequence, and use the license plate as the theoretical license plate of the ETC gantry.

5. According to the method for calculating the recognition accuracy of the ETC gantry snapshot camera according to claim 2, it is characterized in that: To calculate the single-station path, first take the toll station entrance record time as the starting time, and search for the earliest recorded license plate after the vehicle enters the highway on two possible paths, and use the path where this record is located as the vehicle's driving path; Based on this driving path, find the ETC gantries that the vehicle passes through in sequence; Based on the driving path, find the ETC gantries that the vehicle passes through in sequence, and use the license plate as the theoretical license plate of the ETC gantry.

6. According to the method for calculating the recognition accuracy of the ETC gantry snapshot camera according to claim 2, it is characterized in that: Calculate the entry and exit paths, referring to the calculation of the single-station path; based on the driving path, find the ETC gantries that the vehicle passes through in sequence, and use the license plate as the theoretical license plate of the ETC gantry.

7. The method for calculating the recognition accuracy of the ETC gantry snapshot camera according to claim 2 is characterized in that: According to the out-of-town vehicle path, single station path, and entry and exit paths, the theoretical license plate of each ETC gantry is obtained, and it is compared with the recognized license plate of the gantry, and the accuracy is as follows: Taking the theoretical license plate in the identified license plate as the numerator and the theoretical license plate as the denominator, calculate the accuracy rate: Taking the theoretical license plate in the recognized license plate as the numerator and the recognized license plate as the denominator, calculate the actual shooting accuracy:

Images