Matching server, matching system, anomaly detection method, and program
The matching server system addresses the challenge of detecting dirt-induced accuracy drops in license plate readers by comparing license plate information across toll points, ensuring efficient operation without extra hardware.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- MITSUBISHI HEAVY IND MACHINERY SYST LTD
- Filing Date
- 2023-02-17
- Publication Date
- 2026-05-07
AI Technical Summary
Existing license plate readers on main roads and ramps suffer from decreased reading accuracy due to dirt accumulation, which is difficult to detect without adding additional detection devices, leading to increased costs.
A matching server system that utilizes exit, entrance, and intermediate license plate readers to compare and match license plate information across different toll points, determining abnormality in intermediate readers based on agreement rates without requiring additional detection devices.
Enables detection of dirt-induced reading accuracy decreases in intermediate license plate readers by analyzing matching rates, preventing disruptions and maintaining system operation without additional hardware.
Smart Images

Figure 0007854951000001 
Figure 0007854951000002 
Figure 0007854951000003
Abstract
Description
Technical Field
[0001] The present disclosure relates to a matching server, a matching system, an abnormality determination method, and a program.
Background Art
[0002] At the toll gates at the entrances and exits of toll roads, number plate reading devices are installed for use in identifying and managing the vehicle types of vehicles traveling on the toll roads. The number plate reading device includes a camera that photographs the number plate of a vehicle, lighting that irradiates light near the number plate, and an image processing device that performs number plate reading processing (such as OCR processing) on the image captured by the camera.
[0003] In recent years, in addition to toll gates, a mechanism has been considered for installing number plate reading devices at ramps connecting to the main line and other toll roads to identify the routes taken by each vehicle.
[0004] Also, due to reasons such as dirt adhering to the lens or lighting of the camera, the reading accuracy of the number plate reading device may decrease.
[0005] Generally, at a toll gate, the camera and lighting of the number plate reading device are arranged at the side of the road (roadside) according to the height of the number plate. Since the number plate reading device at the toll gate is in a position that is relatively easy for people to reach, it can be cleaned regularly to suppress a decrease in reading accuracy. Also, when a roof is provided at the toll gate, it is less affected by dirt caused by the weather (such as rain or snow), so the decrease in reading accuracy is suppressed accordingly.
[0006] On the other hand, on main roads and ramps, the license plate reading cameras and lights are mounted on poles or gantry, capturing images of license plates and illuminating them from above the road. Because the license plate reading devices on main roads and ramps are located in positions that are difficult for people to reach, they are cleaned less frequently than those at toll booths. In addition, main roads and ramps often lack roofs, making them more susceptible to dirt and grime from weather conditions compared to toll booths. Therefore, it can be difficult to prevent a decrease in reading accuracy due to dirt on license plate reading devices on main roads and ramps.
[0007] Since dirt on the lenses and lighting does not cause changes in the electrical characteristics of the license plate reader, it is difficult to use physical anomaly detection methods that utilize changes in internal voltage and current values, which are common anomaly detection methods. Therefore, it is necessary to implement a mechanism to detect dirt on the lenses and lighting separately from hardware-based anomaly detection methods.
[0008] For example, Patent Document 1 describes a dirt detection device that compares the signal outputs of a first light receiver installed in close contact with the cover (front globe) of a light fixture and a second light receiver installed at a distance from the cover, and detects dirt on the cover of a light fixture when the signal output of the second light receiver decreases compared to that of the first light receiver. [Prior art documents] [Patent Documents]
[0009] [Patent Document 1] Japanese Patent Application Publication No. 63-40203 [Overview of the project] [Problems that the invention aims to solve]
[0010] However, since there are numerous license plate readers installed on main roads and ramps, adding a dirt detection device would significantly increase costs. Therefore, there is a need for technology that can detect abnormalities in license plate readers (decreased reading accuracy) without adding such detection devices.
[0011] The purpose of this disclosure is to provide a matching server, a matching system, an abnormality determination method, and a program that can detect abnormalities in a license plate reader without adding any additional components such as detection devices. [Means for solving the problem]
[0012] According to one aspect of the present disclosure, the matching server includes: an exit vehicle information acquisition unit that acquires exit vehicle information including license plate information read from a vehicle exiting the toll road by an exit license plate reader installed at an exit toll gate of the toll road; an entry vehicle information acquisition unit that acquires entry vehicle information that matches the exit vehicle information from entry vehicle information including license plate information read from a vehicle entering the toll road by an entry license plate reader installed at an entrance toll gate of the toll road; a passing vehicle information acquisition unit that acquires passing vehicle information with the highest agreement rate with the exit vehicle information or the entry vehicle information from passing vehicle information including license plate information read from a vehicle passing a predetermined passing detection position on the toll road by an intermediate license plate reader installed at a predetermined passing detection position on the toll road; and a determination unit that determines whether or not there is an abnormality in the intermediate license plate reader based on the agreement rate.
[0013] According to one aspect of the present disclosure, the matching system comprises an exit license plate reader installed at an exit toll gate of a toll road, an entrance license plate reader installed at an entrance toll gate of the toll road, an intermediate license plate reader installed at a predetermined passage detection position on the toll road, and the matching server described above.
[0014] According to one aspect of the present disclosure, the abnormality determination method includes the steps of: acquiring exit vehicle information including license plate information read from a vehicle exiting the toll road by an exit license plate reader installed at an exit toll gate of the toll road; acquiring entering vehicle information that matches the exit vehicle information from entering vehicle information including license plate information read from a vehicle entering the toll road by an entrance license plate reader installed at an entrance toll gate of the toll road; acquiring passing vehicle information that has the highest agreement rate with the exit vehicle information or the entering vehicle information from passing vehicle information including license plate information read from a vehicle passing the passing detection position by an intermediate license plate reader installed at a predetermined passing detection position of the toll road; and determining whether or not there is an abnormality in the intermediate license plate reader based on the agreement rate.
[0015] According to one aspect of the present disclosure, the program causes a matching server to perform the following steps: acquire exit vehicle information, including license plate information read from a vehicle exiting the toll road by an exit license plate reader installed at an exit toll gate of the toll road; acquire entering vehicle information, including license plate information read from a vehicle entering the toll road by an entrance license plate reader installed at an entrance toll gate of the toll road, that matches the exit vehicle information; acquire passing vehicle information, including license plate information read from a vehicle passing a predetermined passing detection position on the toll road by an intermediate license plate reader installed at a predetermined passing detection position on the toll road, that has the highest matching rate with the exit vehicle information or the entering vehicle information; and determine whether or not there is a malfunction in the intermediate license plate reader based on the matching rate. [Effects of the Invention]
[0016] According to the above embodiment, it is possible to detect abnormalities in the license plate reader without adding any additional components such as detection devices. [Brief explanation of the drawing]
[0017] [Figure 1] It is a schematic diagram showing the overall configuration of a matching system according to an embodiment. [Figure 2] It is a block diagram showing the functional configuration of a matching system according to an embodiment. [Figure 3] It is a graph showing an example of the reading accuracy of a license plate reading device according to an embodiment. [Figure 4] It is a graph showing an example of the matching rate of the reading result according to an embodiment. [Figure 5] It is a sequence diagram showing an example of the processing of a matching system according to an embodiment. [Figure 6] It is a diagram showing an example of entering vehicle information, passing vehicle information, and exiting vehicle information according to an embodiment. [Figure 7] It is a flowchart showing an example of matching processing according to an embodiment. [Figure 8] It is a diagram for explaining matching processing according to an embodiment.
Mode for Carrying Out the Invention
[0018] (Overall Configuration of Matching System) Hereinafter, embodiments will be described in detail with reference to the drawings. FIG. 1 is a schematic diagram showing the overall configuration of a matching system according to an embodiment. As shown in FIG. 1, the matching system 1 includes a matching server 10, an entrance license plate reading device 21, an intermediate license plate reading device 22, and an exit license plate reading device 23.
[0019] The entrance license plate reading device 21 is installed on the roadside of the entrance tollgate of the toll road and reads the license plate of the vehicle entering the toll road from the entrance tollgate.
[0020] The intermediate license plate reader 22 is installed on the gantry at a predetermined passage detection position on the toll road and reads the license plate of a vehicle passing through the passage detection position. The passage detection position can be set at any location on the main road or ramp. For example, as shown in Figure 1, the passage detection position can be provided on at least one or both of ring roads A and B. Alternatively, the passage detection position may be provided on a ramp connecting different toll roads.
[0021] The exit license plate reader 23 is installed on the roadside at the exit toll booth of a toll road and reads the license plates of vehicles exiting the toll road from the exit toll booth.
[0022] The license plate information read by each license plate reader 21, 22, and 23 includes textual information such as place names, classification numbers, and serial numbers written on the license plates. The matching server 10 obtains entering vehicle information D1, which includes the license plate information read by the entrance license plate reader 21; passing vehicle information D2, which includes the license plate information read by the intermediate license plate reader 22; and exiting vehicle information D3, which includes the license plate information read by the exit license plate reader 23.
[0023] The matching server 10 matches the same information between entering vehicle information D1, passing vehicle information D2, and exiting vehicle information D3 from the license plate information read moment by moment by each license plate reader 21, 22, and 23. In this embodiment, the matching server 10 also determines whether there is an abnormality (decreased reading accuracy) in the intermediate license plate reader 22 based on the license plate information matching results.
[0024] Furthermore, the matching system 1 also includes an image server 30. The image server 30 stores the passing vehicle information D2 read by the intermediate license plate reader 22. The matching server 10 obtains the passing vehicle information D2 from the intermediate license plate reader 22 via the image server 30. In other embodiments, the matching server 10 may obtain the passing vehicle information D2 directly from the intermediate license plate reader 22 without going through the image server 30.
[0025] Figure 1 shows a simplified example with one entrance tollbooth, one pass detection point, and one exit tollbooth. However, in reality, there are multiple entrance tollbooths, pass detection points, and exit tollbooths. Therefore, at least one license plate reader 21, 22, 23 is installed at each of the multiple entrance tollbooths, pass detection points, and exit tollbooths. The matching system 1 matches the license plate information read at each location to determine whether there is a malfunction in each of the multiple intermediate license plate readers 22.
[0026] (Functional configuration of license plate reader) Figure 2 is a block diagram showing the functional configuration of a matching system according to one embodiment. As shown in Figure 2, the entrance license plate reading device 21 comprises a camera 200, a lens 201, a light 202, an IF (interface) board 203, and an image processing device 204. The light 202 is, for example, an LED light, which illuminates the area around the vehicle's license plate, enabling the camera 200 to capture the license plate more clearly. The image processing device 204 reads the license plate information by performing predetermined image processing (license plate extraction processing, OCR processing, etc.) on the image captured by the camera 200. The image processing device 204 may also output signals to the camera 200 and lens 201 instructing them on adjustment parameters such as exposure and aperture based on the image captured by the camera 200. The IF board 203 facilitates the exchange of images and signals between the camera 200, lens 201, and light 202 and the image processing device 204. The image processing device 204 of the entrance license plate reading device 21 transmits the read license plate information (entering vehicle information D1) to the matching server 10.
[0027] The intermediate license plate reader 22 and the exit license plate reader 23 have the same functional configuration as the entrance license plate reader 21. The image processing device 204 of the intermediate license plate reader 22 transmits the read license plate information (passing vehicle information D2) to the matching server 10 via the image server 30. The image processing device 204 of the exit license plate reader 23 transmits the read license plate information (exiting vehicle information D3) to the matching server 10.
[0028] (Matching server functional configuration) As shown in Figure 2, the matching server 10 includes a processor 11, memory 12, storage 13, and a communication interface 14.
[0029] The processor 11 operates according to a predetermined program, performing functions as an exiting vehicle information acquisition unit 110, an entering vehicle information acquisition unit 111, a passing vehicle information acquisition unit 112, and a determination unit 113.
[0030] The exiting vehicle information acquisition unit 110 acquires the exiting vehicle information D3, which is the license plate information read by the exiting license plate reader 23.
[0031] The entering vehicle information acquisition unit 111 acquires the entering vehicle information D1, which is the license plate information read by the entrance license plate reader 21, and stores it in the storage 13. The entering vehicle information acquisition unit 111 also searches the storage 13 for the entering vehicle information D1 that matches the exit vehicle information D3 acquired by the exit vehicle information acquisition unit 110, and retrieves it.
[0032] The passing vehicle information acquisition unit 112 acquires from the image server 30 the passing vehicle information D2 that has the highest match rate with the exiting vehicle information D3 or the entering vehicle information D1, from among the passing vehicle information D2 which is license plate information read by the intermediate license plate reader 22. In other embodiments, the passing vehicle information acquisition unit 112 may directly acquire the passing vehicle information D2 from the intermediate license plate reader 22 and store it in the storage 13. In this case, the passing vehicle information acquisition unit 112 searches for and acquires the passing vehicle information D2 that has the highest match rate with the exiting vehicle information D3 or the entering vehicle information D1 from among the passing vehicle information D2 stored in the storage 13.
[0033] The determination unit 113 determines whether there is an abnormality in the intermediate license plate reader 22 that read the passing vehicle information D2, based on the agreement rate between the passing vehicle information D2 acquired by the passing vehicle information acquisition unit 112 and the exiting vehicle information D3 or entering vehicle information D1.
[0034] The predetermined program executed by the processor 11 is stored on a computer-readable recording medium. A computer-readable recording medium refers to a magnetic disk, magneto-optical disk, CD-ROM, DVD-ROM, semiconductor memory, etc. Alternatively, this computer program may be distributed to a computer via a communication line, and the computer that receives the distribution may execute the program. Furthermore, this program may be intended to implement only a part of the functions described above. Moreover, it may be a program that can implement the above functions in combination with a program already recorded in the computer system, a so-called differential file (differential program).
[0035] Memory 12 has a memory area necessary for the operation of the processor 11.
[0036] Storage 13 is a so-called auxiliary storage device, such as an HDD (Hard Disk Drive) or SSD (Solid State Drive). Data acquired, generated, and referenced by each part of the processor 11 during processing is stored in storage 13.
[0037] Communication IF14 is an interface for sending and receiving various data and control signals between the license plate readers 21, 22, and 23 and the image server 30.
[0038] (Regarding the reading accuracy of license plate readers) Figure 3 is a graph showing an example of the reading accuracy of a license plate reader according to one embodiment. Figure 3 shows an example of the degree of decrease in reading accuracy over time between the entrance license plate reader 21 or the exit license plate reader 23 and the intermediate license plate reader 22. In Figure 3, the vertical axis represents reading accuracy (%), and the horizontal axis represents elapsed time.
[0039] Graph G11 shows an example of the degree of decrease in reading accuracy of the entrance license plate reader 21 or the exit license plate reader 23. Graphs G12 and G13 both show examples of the degree of decrease in reading accuracy of the intermediate license plate reader 22.
[0040] As described above, the entrance license plate reader 21 and the exit license plate reader 23 are protected from dirt caused by rain and snow by the roof of the toll booth. In addition, maintenance such as cleaning and replacement of consumables is easy. Therefore, as shown in graph G11 of Figure 3, the entrance license plate reader 21 and the exit license plate reader 23 can maintain a high reading accuracy (close to 100%) even after time has passed.
[0041] The intermediate license plate reader 22 is often installed in open locations such as on the main road or on ramps, making it susceptible to dirt and grime from rain and snow. Furthermore, because the intermediate license plate reader 22 is installed in high places such as on the gantry, it requires less frequent cleaning and maintenance compared to the entrance license plate reader 21. As a result, as shown in graph G12 of Figure 3, the reading accuracy of the intermediate license plate reader 22 decreases over time due to deterioration (for example, a decrease in illuminance due to the deterioration of the lighting 202). In addition, if dirt accumulates on the lens 201 or lighting 202 of the intermediate license plate reader 22, the reading accuracy decreases significantly from the moment the dirt accumulates, as shown in graph G13 of Figure 3.
[0042] The intermediate license plate reader 22 alone cannot determine whether the read license plate information (reading result) is correct or incorrect. Therefore, it is difficult to determine whether the reading accuracy has decreased based solely on the reading result of the intermediate license plate reader 22. On the other hand, the entrance license plate reader 21 and the exit license plate reader 23 can maintain a relatively high reading accuracy, as shown in the example in Figure 3. Therefore, in the matching system 1 according to this embodiment, the reading result of the entrance license plate reader 21 or the exit license plate reader 23 is used as the correct data and compared with the reading result of the intermediate license plate reader 22 to determine whether the reading accuracy of the intermediate license plate reader 22 has decreased.
[0043] Figure 4 is a graph showing an example of the agreement rate of reading results according to one embodiment. Figure 4 shows an example of the degree of decrease in the agreement rate between the reading results (correct data) of the entrance license plate reader 21 or the exit license plate reader 23 and the reading results of the intermediate license plate reader 22, with respect to the passage of time. In Figure 4, the vertical axis represents the agreement rate (%), and the horizontal axis represents the passage of time. The agreement rate is, for example, the ratio of the number of characters that matched between the correct data and the reading results of the intermediate license plate reader 22 to the total number of characters included in the reading result (total number of characters written on the license plate) (i.e., agreement rate = number of matching characters / total number of characters). Graph G22 corresponds to graph G12 in Figure 3 and shows how the agreement rate gradually decreases in accordance with the decrease in reading accuracy due to the aging deterioration of the intermediate license plate reader 22. Also, graph G23 corresponds to graph G13 in Figure 3 and shows how the agreement rate decreases significantly from the point when dirt adheres to the intermediate license plate reader 22. In this embodiment, the matching system 1 detects that the reading accuracy of the intermediate license plate reader 22 has decreased due to dirt if the matching rate has significantly decreased compared to the trend of decreasing matching rate due to aging (graph G22).
[0044] (Processing flow of the matching system) Next, we will explain in detail the process by which the matching system 1 detects whether or not there is a decrease in the reading accuracy of the intermediate license plate reader 22.
[0045] Figure 5 is a sequence diagram showing an example of the processing of a matching system according to one embodiment. Figure 5 shows an example of the processing of the matching system 1 when a vehicle travels on a toll road. First, when the vehicle enters the toll road from the entrance toll gate, the entrance license plate reader 21 photographs the vehicle's license plate and reads the license plate information (step S01). Since these processes are based on known technology, a detailed explanation is omitted.
[0046] Furthermore, the entrance license plate reader 21 transmits the entered vehicle information D1, which includes the read license plate information and the reading time, to the matching server 10 (step S02). The entered vehicle information D1 is also accompanied by information that identifies the entrance tollbooth where the entrance license plate reader 21 is installed (for example, a tollbooth ID). The entrance license plate reader 21 may also transmit the image used for the reading process to the matching server 10 along with the entered vehicle information D1.
[0047] When the matching server 10's vehicle entry information acquisition unit 111 receives vehicle entry information D1 from the entrance license plate reader 21, it records and stores it in the storage 13 (step S03).
[0048] Figure 6 shows an example of entering vehicle information, passing vehicle information, and exiting vehicle information according to one embodiment. As shown in Figure 6, the vehicle entry information acquisition unit 111 may record the vehicle entry information D1 in the vehicle entry information table D10 for each entrance toll gate, based on the toll gate ID attached to the vehicle entry information D1.
[0049] Next, let's assume the vehicle has passed the detection point on Loop Line A (Figure 1). Then, the intermediate license plate reader 22 installed at the detection point on Loop Line A takes a photograph of the vehicle's license plate and reads the license plate information (Step S04).
[0050] Furthermore, the intermediate license plate reader 22 transmits passing vehicle information D2, which includes the read license plate information and the reading time, to the image server 30 (step S05). In addition, the passing vehicle information D2 is accompanied by information that can identify the passing detection location (ring road A) where the intermediate license plate reader 22 is installed (for example, a detection location ID). The intermediate license plate reader 22 may also transmit the image used for the reading process to the image server 30 along with the passing vehicle information D2.
[0051] When the image server 30 receives passing vehicle information D2 from the intermediate license plate reader 22, it records and stores it in storage (not shown) (step S06). Alternatively, as shown in Figure 6, the image server 30 may record the passing vehicle information D2 in a passing vehicle information table D20 for each passing detection position, based on the detection position ID attached to the passing vehicle information D2.
[0052] Next, let's assume the vehicle exits the toll road through the exit toll gate. Then, the exit license plate reader 23 installed at the exit toll gate takes a photograph of the vehicle's license plate and reads the license plate information (step S07).
[0053] Furthermore, the exit license plate reader 23 transmits exit vehicle information D3, which includes the read license plate information and the time of reading, to the matching server 10 (step S08). In addition, the exit vehicle information D3 is accompanied by information that identifies the exit toll gate where the exit license plate reader 23 is installed (toll gate ID). The exit license plate reader 23 may also transmit the image used for the reading process to the matching server 10 along with the exit vehicle information D3.
[0054] When the exit vehicle information acquisition unit 110 of the matching server 10 receives exit vehicle information D3 from the exit license plate reader 23, it records and stores it in the storage 13 (step S09). As shown in Figure 6, the exit vehicle information acquisition unit 110 may also record the exit vehicle information D3 in the exit vehicle information table D30 for each exit toll gate based on the toll gate ID attached to the exit vehicle information D3.
[0055] Upon obtaining the exit vehicle information D3, the matching server 10 executes the matching process (step S10).
[0056] Figure 7 is a flowchart showing an example of a matching process according to one embodiment. Figure 8 is a diagram illustrating a matching process according to one embodiment. Here, we will explain the details of the matching process S10 of the matching server 10 with reference to Figures 7 and 8.
[0057] First, the entering vehicle information acquisition unit 111 searches for and acquires an entering vehicle information D1 that matches the exiting vehicle information D3 acquired by the exiting vehicle information acquisition unit 110 from among the multiple entering vehicle information D1 recorded in the entering vehicle information table D10 (step S101).
[0058] As shown in Figure 8, the reading time included in the exit vehicle information D3 acquired by the exit vehicle information acquisition unit 110 is denoted as T3, and the license plate information as X3. The entry vehicle information acquisition unit 111 extracts the entry vehicle information D1 recorded in the entry vehicle information table D10 of each of the multiple entrance toll gates as search target data TD1 if it was read before the reading time T3 of the exit vehicle information D3 (in the example in Figure 8, "11:00").
[0059] Furthermore, the entering vehicle information acquisition unit 111 searches for and acquires entering vehicle information D1 from the extracted search target data TD1, for example, in order of the most recent reading time, and which has license plate information that matches the license plate information X3 of the exiting vehicle information D3. The reading time of the acquired entering vehicle information D1 is denoted as T1, and the license plate information as X1.
[0060] The incoming vehicle information acquisition unit 111 may also refer to the toll gate data pre-recorded in the storage 13 and extract the search target data TD1 only from the incoming vehicle information table D10 of the entrance toll gate located upstream of the exit toll gate from which the exit vehicle information D3 was read. By narrowing down the search target data in this way, the search process can be sped up.
[0061] Next, the passing vehicle information acquisition unit 112 extracts the passing vehicle information D2 as search target data TD2, which was read during the period after the reading time T1 of the entering vehicle information D1 acquired in step S101 (10:10 in the example of Figure 8) and before the reading time T3 of the exiting vehicle information D3 (11:00 in the example of Figure 8) (step S102).
[0062] In this case, the passing vehicle information acquisition unit 112 may refer to the toll gate data pre-recorded in the storage 13 and extract the search target data TD2 only from the passing vehicle information table D20 of passing detection locations located downstream of the entrance toll gate from which the entering vehicle information D1 was read, and upstream of the exit toll gate from which the exiting vehicle information D3 was read. By narrowing down the search target data in this way, the search process can be sped up.
[0063] Furthermore, the passing vehicle information acquisition unit 112 calculates the percentage of agreement (%) between each of the license plate information entries in the passing vehicle information D2 extracted as the search target data TD2 and the license plate information X3 of the exiting vehicle information D3 (step S103). The passing vehicle information acquisition unit 112 may also calculate the percentage of agreement between the license plate information X1 of the entering vehicle information D1 and the passing vehicle information D1.
[0064] Next, the passing vehicle information acquisition unit 112 acquires the passing vehicle information D2 with the highest matching rate (for example, 70%) as data that matches the exiting vehicle information D3 (or entering vehicle information D1) (step S104).
[0065] Furthermore, if there are multiple passing detection locations (intermediate license plate readers 22) in the section from the entrance toll gate to the exit toll gate, the passing vehicle information acquisition unit 112 may acquire passing vehicle information D2 for each passing detection location. Also, if there is no passing vehicle information D2 in the search target data TD2 for a certain passing detection location with a matching rate equal to or greater than the matching threshold, the passing vehicle information acquisition unit 112 determines that the vehicle has not passed through that passing detection location. The passing vehicle information acquisition unit 112 does not acquire passing vehicle information D2 from the search target data TD2 for passing detection locations that it has determined the vehicle has not passed through. As a result, the passing vehicle information acquisition unit 112 can acquire only the passing vehicle information D2 read at the passing detection locations through which the vehicle actually passed.
[0066] Although not shown in the diagram, the passing vehicle information acquisition unit 112 may also terminate the matching process S10 if there are no passing detection locations in the section from the entrance toll gate to the exit toll gate, assuming that no matching passing vehicle information D2 exists.
[0067] The matching process S10 allows for the acquisition of entering vehicle information D1, passing vehicle information D2, and exiting vehicle information D3, each associated with a single vehicle. This matching result is recorded in the storage 13 and may be used, for example, to identify the vehicle's travel route. In this embodiment, this matching result is also used to determine whether or not there is a decrease in the reading accuracy of the intermediate license plate reader 22. This determination process will be explained with reference to the sequence shown in Figure 5.
[0068] The determination unit 113 of the matching server 10 calculates the matching rate of the intermediate license plate reader 22 over a certain period (step S11). For example, the determination unit 113 extracts the matching results of the intermediate license plate reader 22 on ring road A (Figure 1) over a certain period from the present to the past (for example, one month) from the matching results recorded in the storage 13, and calculates the average value of the matching rate.
[0069] Next, the determination unit 113 determines that there is a problem with the intermediate license plate reader 22 on ring road A, that is, that the reading accuracy has decreased due to dirt, if the agreement rate over a certain period falls below a predetermined abnormality determination threshold (for example, 70%) (step S12; YES). The abnormality determination threshold may be arbitrarily changed based on the decreasing trend of the agreement rate as illustrated in Figure 4, or environmental conditions (trends in weather and sunlight conditions at the vehicle detection location, presence or absence of structures that block rain or snow such as roofs, whether it is a place with a high volume of traffic of vehicles with the same place name or classification number, etc.).
[0070] Furthermore, if the matching rate falls below the abnormality detection threshold (step S12; YES), the determination unit 113 notifies the image server 30 of an abnormality (decreased reading accuracy) in the intermediate license plate reader 22 of ring road A (step S13). Upon receiving the abnormality notification, the image server 30 sends an abnormality notification to the intermediate license plate reader 22 of ring road A, which is the source of the abnormality (step S15).
[0071] Furthermore, the determination unit 113 may also send abnormality notifications to higher-level servers (not shown) (for example, management servers of toll road operators or maintenance companies) in addition to the image server 30. This allows, for example, maintenance companies to quickly clean or maintain the intermediate license plate reader 22 on ring road A. This prevents disruptions to the operation of the matching system 1 (for example, identification of vehicle travel routes) due to a decrease in the reading accuracy of the intermediate license plate reader 22. Alternatively, even if disruptions occur to the operation of the matching system 1, they can be restored in a short period of time.
[0072] Furthermore, if the agreement rate over a certain period is above a predetermined abnormality determination threshold (step S12; NO), the determination unit 113 determines that the intermediate license plate reader 22 on ring road A is functioning normally and terminates the determination process for this intermediate license plate reader 22 (step S14).
[0073] Furthermore, if multiple intermediate license plate readers 22 are installed on a toll road, the determination unit 113 performs the same process in steps S11 to S14 for each intermediate license plate reader.
[0074] Furthermore, Figure 5 illustrates an example in which the determination unit 113 calculates the agreement rate over a certain period and determines that the intermediate license plate reader 22 is abnormal if the agreement rate falls below an abnormality determination threshold, but the system is not limited to this example. In other embodiments, the determination unit 113 may calculate the mismatch rate over a certain period and determine that the intermediate license plate reader 22 is abnormal if the mismatch rate exceeds an abnormality determination threshold (for example, 30%).
[0075] (Effect, Action) As described above, the matching server 10 according to this embodiment includes: an exit vehicle information acquisition unit 110 that acquires exit vehicle information D3 including license plate information read by the exit license plate reader 23; an entry vehicle information acquisition unit 111 that acquires entry vehicle information D1 from among entry vehicle information D1 including license plate information read by the entrance license plate reader 21 that matches the exit vehicle information D3; a passing vehicle information acquisition unit 112 that acquires passing vehicle information D2 from among passing vehicle information D2 including license plate information read by the intermediate license plate reader 22 that has the highest match rate with the exit vehicle information D3 or entry vehicle information D1; and a determination unit 113 that determines whether or not there is an abnormality (decreased reading accuracy) in the intermediate license plate reader 22 based on the match rate.
[0076] In this way, the matching server 10 can determine whether or not there is an abnormality due to dirt (a decrease in reading accuracy), which is difficult to detect by the hardware abnormality detection method of the intermediate license plate reader 22, based on the reading results of each license plate reader 21, 22, and 23, without adding any detection devices or other configurations to the intermediate license plate reader 22.
[0077] Furthermore, the vehicle information acquisition unit 112 of the matching server 10 extracts vehicle information D2 that was read after the reading time T1 of the entering vehicle information D1 and before the reading time T3 of the exiting vehicle information D3, and acquires the vehicle information D2 with the highest matching rate among the extracted vehicle information D2.
[0078] In this way, the matching server 10 can narrow down the candidates for the passing vehicle information D2 to be searched based on the reading times of the entering vehicle information D1 and the exiting vehicle information D3. This allows the matching server 10 to speed up the search process for passing vehicle information D2.
[0079] Furthermore, the passing vehicle information acquisition unit 112 extracts passing vehicle information D2 read by intermediate license plate readers 22 installed in the section from the entrance toll gate where the entrance license plate reader 21 that reads the entering vehicle information D1 is installed, to the exit toll gate where the exit license plate reader 23 that reads the exiting vehicle information D3 is installed, and acquires the passing vehicle information D2 with the highest matching rate among the extracted passing vehicle information D2.
[0080] In this way, the matching server 10 can narrow down the candidates for the passing vehicle information D2 to be searched based on the locations where the entering vehicle information D1 and exiting vehicle information D3 were read. This allows the matching server 10 to speed up the search process for passing vehicle information D2.
[0081] Furthermore, the passing vehicle information acquisition unit 112 calculates the agreement rate with the entering vehicle information D1 or exiting vehicle information D3 for each of the multiple passing vehicle information D2s, and acquires the passing vehicle information D2 that has the highest agreement rate and is equal to or greater than a predetermined agreement threshold.
[0082] In this way, if there is no passing vehicle information D2 that exceeds the matching threshold, the matching server 10 can determine that the vehicle did not pass the installation location (passing detection location) of the intermediate license plate reader 22 that reads this passing vehicle information D2, thereby suppressing the matching with incorrect passing vehicle information D2.
[0083] Furthermore, the matching system 1 according to this embodiment includes an exit license plate reader 23 installed at the exit toll gate of the toll road, an entrance license plate reader 21 installed at the entrance toll gate of the toll road, an intermediate license plate reader 22 installed at a predetermined passage detection position on the toll road, and the matching server 10 described above.
[0084] In this way, the matching system 1 can determine whether or not there is an abnormality due to dirt (a decrease in reading accuracy), which is difficult to detect by the hardware abnormality detection method of the intermediate license plate reader 22, based on the reading results of each license plate reader 21, 22, and 23, without adding any detection devices or other configurations to the intermediate license plate reader 22.
[0085] As described above, embodiments relating to this disclosure have been explained, but these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents.
[0086] <Note> The matching server, matching system, anomaly detection method, and program described in the above-described embodiment can be understood, for example, as follows.
[0087] (1) According to the first embodiment, the matching server 10 includes: an exit vehicle information acquisition unit 110 that acquires exit vehicle information D3, which includes license plate information read from a vehicle exiting the toll road by an exit license plate reader 23 installed at the exit toll gate of the toll road; an entry vehicle information acquisition unit 111 that acquires entry vehicle information D1, which includes license plate information read from a vehicle entering the toll road by an entry license plate reader 21 installed at the entrance toll gate of the toll road, and that matches the exit vehicle information D3; a passing vehicle information acquisition unit 112 that acquires passing vehicle information D2, which includes license plate information read from a vehicle passing a passing detection position by an intermediate license plate reader 22 installed at a predetermined passing detection position on the toll road, and that has the highest matching rate with the exit vehicle information D3 or the entry vehicle information D1; and a determination unit 113 that determines whether or not there is an abnormality in the intermediate license plate reader 22 based on the matching rate.
[0088] In this way, the matching server 10 can determine whether or not there is an abnormality due to dirt (a decrease in reading accuracy), which is difficult to detect by the hardware abnormality detection method of the intermediate license plate reader 22, based on the reading results of each license plate reader 21, 22, and 23, without adding any detection devices or other configurations to the intermediate license plate reader 22.
[0089] (2) According to the second embodiment, in the matching server 10 according to the first embodiment, the passing vehicle information acquisition unit 112 extracts passing vehicle information D2 that was read after the reading time T1 of the entering vehicle information D1 and before the reading time T3 of the exiting vehicle information D3, and acquires the passing vehicle information D2 with the highest matching rate among the extracted passing vehicle information D2.
[0090] In this way, the matching server 10 can narrow down the candidates for the passing vehicle information D2 to be searched based on the reading times of the entering vehicle information D1 and the exiting vehicle information D3. This allows the matching server 10 to speed up the search process for passing vehicle information D2.
[0091] (3) According to the third embodiment, in the matching server 10 according to the first or second embodiment, the passing vehicle information acquisition unit 112 extracts passing vehicle information D2 read by intermediate license plate readers 22 installed in the section from the entrance toll gate where the entrance license plate reader 21 that reads the entering vehicle information D1 is installed to the exit toll gate where the exit license plate reader 23 that reads the exiting vehicle information D3 is installed, and acquires the passing vehicle information D2 with the highest matching rate among the extracted passing vehicle information D2.
[0092] In this way, the matching server 10 can narrow down the candidates for the passing vehicle information D2 to be searched based on the locations where the entering vehicle information D1 and exiting vehicle information D3 were read. This allows the matching server 10 to speed up the search process for passing vehicle information D2.
[0093] (4) According to the fourth embodiment, in the matching server 10 according to any one of the first to third embodiments, the passing vehicle information acquisition unit 112 calculates the agreement rate with the exiting vehicle information D3 or the entering vehicle information D1 for each of the multiple passing vehicle information D2s, and acquires the passing vehicle information D2 that has the highest agreement rate and is equal to or greater than a predetermined agreement threshold.
[0094] In this way, if there is no passing vehicle information D2 that exceeds the matching threshold, the matching server 10 can determine that the vehicle did not pass the installation location (passing detection location) of the intermediate license plate reader 22 that reads this passing vehicle information D2, thereby suppressing the matching with incorrect passing vehicle information D2.
[0095] (5) According to the fifth embodiment, the matching system 1 comprises an exit license plate reader 23 installed at the exit toll gate of the toll road, an entrance license plate reader 21 installed at the entrance toll gate of the toll road, an intermediate license plate reader 22 installed at a predetermined passage detection position on the toll road, and a matching server 10 according to any one of the first to fourth embodiments.
[0096] In this way, the matching system 1 can determine whether or not there is an abnormality due to dirt (a decrease in reading accuracy), which is difficult to detect by the hardware abnormality detection method of the intermediate license plate reader 22, based on the reading results of each license plate reader 21, 22, and 23, without adding any detection devices or other configurations to the intermediate license plate reader 22.
[0097] (6) According to the sixth embodiment, the matching method includes the steps of: acquiring exit vehicle information D3, which includes license plate information read from a vehicle exiting the toll road by an exit license plate reader 23 installed at the exit toll gate of the toll road; acquiring entry vehicle information D1, which includes license plate information read from a vehicle entering the toll road by an entrance license plate reader 21 installed at the entrance toll gate of the toll road, and which matches the exit vehicle information D3; acquiring entry vehicle information D2, which includes license plate information read from a vehicle passing through a passage detection position by an intermediate license plate reader 22 installed at a predetermined passage detection position on the toll road, and which has the highest matching rate with the exit vehicle information D3 or entry vehicle information D1; and determining whether or not there is a malfunction in the intermediate license plate reader 22 based on the matching rate.
[0098] (7) According to the seventh aspect, the program causes the matching server 10 to execute the following steps: acquiring exit vehicle information D3, which includes license plate information read from a vehicle exiting the toll road by an exit license plate reader 23 installed at the exit toll gate of the toll road; acquiring entering vehicle information D1, which includes license plate information read from a vehicle entering the toll road by an entrance license plate reader 21 installed at the entrance toll gate of the toll road, and matches the exit vehicle information D3; acquiring passing vehicle information D2, which includes license plate information read from a vehicle passing a passing detection position by an intermediate license plate reader 22 installed at a predetermined passing detection position on the toll road, and has the highest matching rate with the exit vehicle information D3 or the entering vehicle information D1; and determining whether or not there is an abnormality in the intermediate license plate reader 22 based on the matching rate. [Explanation of symbols]
[0099] 1. Matching System 10 Matching Servers 11 processors 110 Exiting Vehicle Information Acquisition Unit 111 Approaching vehicle information acquisition unit 112 Passing Vehicle Information Acquisition Unit 113 Judgment section 12 memory 13 Storage 14 Communication IF 21 Entrance License Plate Reader 22 Intermediate license plate reader 23. Exit License Plate Reader 200 Cameras 201 Lens 202 Lighting 203 IF board 204 Image Processing Device 30 Image Servers D1 Entry Vehicle Information D2 Passing Vehicle Information D3 Exit Vehicle Information
Claims
1. An exit vehicle information acquisition unit acquires exit vehicle information, including license plate information, from a vehicle exiting the toll road, which is read by an exit license plate reader installed at the exit toll gate of the toll road. An entry vehicle information acquisition unit acquires entry vehicle information, which includes license plate information read from vehicles entering the toll road by an entry license plate reader installed at the entrance toll gate of the aforementioned toll road, and which matches the entry vehicle information with the exit vehicle information. A passing vehicle information acquisition unit acquires passing vehicle information, which includes license plate information read from vehicles passing through a passing detection position by an intermediate license plate reader installed at a predetermined passing detection position on the toll road, and which has the highest agreement rate with the exiting vehicle information or the entering vehicle information. A determination unit that determines whether or not there is an abnormality in the intermediate license plate reader based on the aforementioned matching rate, A matching server equipped with the following features.
2. The aforementioned vehicle information acquisition unit, Extract the passing vehicle information read during the period after the time the entering vehicle information was read and before the time the exiting vehicle information was read. From the extracted passing vehicle information, retrieve the passing vehicle information with the highest matching rate. The matching server according to claim 1.
3. The aforementioned vehicle information acquisition unit, The information of passing vehicles read by intermediate license plate readers installed in the section from the entrance toll gate where the entrance license plate reader that reads the information of entering vehicles is installed to the exit toll gate where the exit license plate reader that reads the information of exiting vehicles is installed is extracted. Obtain the vehicle information with the highest matching rate from the extracted vehicle information. The matching server according to claim 1.
4. The passing vehicle information acquisition unit calculates the agreement rate with the exiting vehicle information or the entering vehicle information for each of the multiple passing vehicle information entries, and acquires the passing vehicle information for which the agreement rate is equal to or greater than a predetermined agreement threshold and which has the highest agreement rate. The matching server according to claim 1.
5. An exit license plate reader installed at the toll booth of a toll road, An entrance license plate reader installed at the entrance toll booth of the aforementioned toll road, An intermediate license plate reader installed at a predetermined passage detection position on the aforementioned toll road, A matching server as described in any one of claims 1 to 4, A matching system equipped with the following features.
6. The matching server obtains exit vehicle information, which includes license plate information read from a vehicle exiting the toll road by an exit license plate reader installed at the exit toll gate of the toll road, The matching server obtains, from among the entering vehicle information, which includes license plate information read from vehicles entering the toll road by an entrance license plate reader installed at the entrance toll gate of the toll road, the entering vehicle information that matches the exiting vehicle information. The matching server obtains, from among the passing vehicle information, which includes license plate information read from vehicles passing through a predetermined passing detection position on the toll road by an intermediate license plate reader installed at the passing detection position, the passing vehicle information with the highest matching rate with the exiting vehicle information or the entering vehicle information. The matching server determines whether or not there is an abnormality in the intermediate license plate reader based on the matching rate, A method for detecting abnormalities in a clutch.
7. The steps include: obtaining exit vehicle information, including license plate information, from a vehicle exiting the toll road, using an exit license plate reader installed at the exit toll gate of the toll road; The steps include: obtaining entering vehicle information, which includes license plate information read from a vehicle entering the toll road by an entrance license plate reader installed at the entrance toll gate of the toll road, and which matches the exiting vehicle information; The steps include: acquiring passing vehicle information, which includes license plate information read from vehicles passing through a designated passing detection position on the toll road by an intermediate license plate reader installed at a predetermined passing detection position on the toll road, from among the passing vehicle information, the passing vehicle information with the highest matching rate with the exiting vehicle information or the entering vehicle information; Based on the aforementioned matching rate, the step of determining whether or not there is an abnormality in the intermediate license plate reader, A program that instructs the matching server to execute a command.
Citation Information
Patent Citations
Fare reception for toll road
JP1986033587A
Fouling detector for lighting fixture
JP1988040203A
Charge collection system, on-vehicle equipment, and charge collection method
JP2003216989A
Traffic lane control system, traffic lane control method, and program
JP2015153383A
Identification device, identification method, and program
JP2023013055A