Parking management system

Abstract

To provide a system for detecting vehicles entering / leaving a management object area based on a photographed image, and grasping the number of vehicles existing in the management object area by performing summarization.SOLUTION: A parking lot management system includes: an imaging device for photographing a parking lot and outputting image information; and an image information processing device. The image information processing device includes image information processing means having: vehicle detection means for detecting a vehicle entering a management object area and a vehicle leaving the management object area based on image information of the management object area of the parking lot by the imaging device so as to output a detection result; and summarization means for adding the number of vehicles entering the management object area upon receiving an output of the vehicle detection means, and summarizing the number of vehicles existing in the management object area by subtracting the number of vehicles leaving the management object area.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] This application relates to a parking lot management system.

Background Art

[0002] In order to grasp the usage status of a parking lot (the vacancy status of parking spaces), image information of parked vehicles is acquired by an imaging device arranged at a high position, and the obtained image information is analyzed.

[0003] Also, in detecting the usage status of parking spaces, the presence or absence of vehicles is discriminated using the contour method or machine learning method of image processing technology. Further, when the image is unclear, clear images with the same usage status of the parking spaces are detected from the captured images and used to suppress noise and improve the detection accuracy (Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] As shown in Patent Document 1, in order to grasp the number of vehicles present in a parking lot, it is necessary to improve the imaging technology for photographing the parking lot and the information processing technology for extracting necessary information from the captured image information. However, when the number of parked vehicles increases, there is a problem that the accuracy of grasping the number of vehicles from the image information decreases because the vehicles behind are hidden by the shadows of the vehicles.

[0006] An object of this application is to provide a parking lot management system that substitutes for a system for grasping the number of vehicles from the image information of parked vehicles with respect to the above - mentioned problems. [Means for solving the problem]

[0007] The parking management system of the present invention comprises an imaging device that photographs a parking lot and outputs image information, and an image information processing device. It was obtained The image information processing device includes a vehicle detection means that detects vehicles entering and leaving the managed area of ​​the parking lot based on image information of the managed area of ​​the parking lot obtained by the imaging device and outputs the detection results, and an aggregation means that receives the output of the vehicle detection means and aggregates the number of vehicles present in the managed area by adding the number of vehicles entering the managed area and subtracting the number of vehicles leaving the managed area. The image information processing means includes an image information analysis means that determines the number of vehicles present in the managed area based on an image captured by the imaging device, the determination of the number of vehicles by the image information analysis means is performed periodically at predetermined time intervals based on an image captured by the imaging device, detection by the vehicle detection means is performed each time a vehicle enters or leaves, and the period of determination of the number of vehicles by the image information analysis means is set to be shorter than the period at which the vehicle detection means outputs the detection result. It is characterized by the following. [Effects of the Invention]

[0008] According to the parking management system of the present invention, a system is provided that can determine the number of vehicles present in a managed area by detecting and aggregating vehicles entering and exiting the managed area based on images captured by an imaging device. [Brief explanation of the drawing]

[0009] [Figure 1] This diagram shows the configuration of the parking management system according to Embodiment 1. [Figure 2] This is a plan view showing an example of a parking lot. [Figure 3] This figure shows an example of an overhead view of the area to be managed in a parking lot. [Figure 4] This is a plan view of the area to be managed in the parking lot. [Figure 5] This is a flowchart showing the aggregation process in Embodiment 1. [Figure 6] This diagram illustrates the aggregation of the number of vehicles using the vehicle detection means of Embodiment 1. [Figure 7] This diagram illustrates the operation of the passage image information processing means in Embodiments 1 and 2. [Figure 8] This diagram shows the configuration of the parking management system according to Embodiment 2. [Figure 9] This is a diagram for explaining the number of vehicles by the image information analysis means of Embodiment 2. [Figure 10] This is a diagram for explaining the number of vehicles by the vehicle detection means and the number of vehicles by the image information analysis means of Embodiment 2. [Figure 11] This is a flowchart of the aggregation means of Embodiment 2. [Figure 12] This is a diagram for explaining the case where the period of the image information analysis means of Embodiment 2 is shortened. [Figure 13] This is a diagram for explaining the first threshold value of Embodiment 3. [Figure 14] This is a flowchart of Embodiment 3. [Figure 15] This is a diagram for explaining the second threshold value of Embodiment 4. [Figure 16] This is a flowchart of Embodiment 4. [Figure 17] This is a diagram for explaining the first threshold value and the second threshold value of Embodiment 4. [Figure 18] This is a diagram for explaining the variation of the second threshold value of Embodiment 5. [Figure 19] This is a hardware configuration diagram for realizing the information processing apparatus of the embodiment.

Embodiments for Carrying Out the Invention

[0010] Hereinafter, embodiments of the parking lot management system according to the present application will be described with reference to the drawings. In each figure, the same or corresponding parts are denoted by the same reference numerals, and redundant descriptions are omitted.

[0011] Embodiment 1. FIG. 1 is a configuration diagram showing a parking lot management system according to Embodiment 1. As shown in FIG. 1, the parking lot management system 1000 includes an imaging device 10 that captures images of the parking lot, and an information processing system 20 that analyzes the image information captured by the imaging device 10 to obtain necessary information.

[0012] The imaging device 10 is mounted in the parking lot at a position higher than the height of the vehicles, and it photographs the entire parking lot and outputs the image information to the information processing system 20. The information processing system 20 understands the usage status of the parking lot by processing the image information output from the imaging device 10.

[0013] The information processing system 20 includes an image information processing device 21 and a management device 22. The image information processing device 21 includes a first image information processing means 210, a second image information processing means 220, and a passage image information processing means 230.

[0014] The first image information processing means 210 includes a vehicle detection means 211 and an aggregation means 212. Similarly, the second image information processing means 220 includes a vehicle detection means 221 and an aggregation means 222. Each of the first image information processing means 210 and the second image information processing means 220 is assigned to process the image information of a predetermined managed area of ​​the parking lot.

[0015] Furthermore, the aisle image information processing means 230 is responsible for processing not only the image information of the parking spaces but also the image information of the aisles 103 of the parking lot, and includes a vehicle identification means 231, a parking position estimation means 232, and a monitoring means 233.

[0016] The operation of the first image information processing means 210 within the image information processing device 21 will be described. Note that the second image information processing means 220 operates similarly, although it handles a different area of ​​data, so its description will be omitted.

[0017] The vehicle detection means 211 receives image information from the imaging device 10, extracts vehicle images contained in the image information, and detects whether a vehicle is entering a parking space in the managed area or leaving the managed area by observing the direction of movement of the vehicle image relative to a predetermined managed area in the parking lot. If a vehicle entering the managed area is detected, it outputs a signal to add, and if a vehicle leaving the managed area is detected, it outputs a signal to subtract.

[0018] Upon receiving the output from the vehicle detection means 211, the aggregation means 212 adds and subtracts from the number of vehicles in the managed area and outputs the aggregated value to the management device 22. Here, vehicle image extraction can be performed with near-perfect accuracy from image information, even if it's only a part of an image, by pre-training the system with images of various vehicles and storing them in memory.

[0019] The management device 22 receives the output from the aggregation means 212 and outputs management information such as information regarding the congestion status of the parking lot. Furthermore, the control device 22 receives image information from the imaging device 10 and controls the direction, angle, magnification, and other aspects of imaging by the imaging device 10.

[0020] Figure 2 shows an example of a parking lot managed by this parking management system 1000. Figure 2 is a plan view of the parking lot, showing the arrangement of the imaging device 10 in relation to the parking lot 100, as well as the parking spaces 101, the managed area 102, the passageway 103, and the vehicles 50 in the parking lot 100.

[0021] When managing the parking lot 100, a plan view as shown in Figure 2 is required. In this embodiment 1, the parking lot 100 is divided into multiple areas, which are designated as the managed area 102. That is, the managed area 102 is an area that combines a predetermined number of parking spaces 101 into one. In the example shown in Figure 2, the areas labeled A, B, C, D, E, F, G, and H are designated as managed areas 102. The passageway 103 is provided so that the vehicle 50 can travel between the managed areas 102.

[0022] Figure 3 shows an example of an overhead view of the managed area 102 of parking lot 100, which was captured by the imaging device 10. Figure 3 shows image information for one frame of a series of overhead images. By observing multiple frames of this overhead image, it is possible to detect vehicles 51 moving out of the managed area 102 and vehicles 52 moving into the managed area 102.

[0023] When the overhead view in Figure 3 is represented as a plan view, it becomes as shown in Figure 4. Here, the vehicles 50 located within the managed area 102 are each located within the parking space 101. The number of these vehicles 50 can be determined by conventional image analysis, but in this embodiment 1, the number is tallied by detecting moving vehicles (vehicles 51 leaving the managed area 102 and vehicles 52 entering the managed area 102) as shown in Figure 4.

[0024] The aggregation process according to this embodiment 1 is performed according to the procedure shown in Figure 5. As shown in Figure 5, first, the vehicle detection means 211 receives image information of the managed area 102 (image information of an overhead view) from the imaging device 10 (step S51). The vehicle detection means 211 then extracts images of vehicles from the image information to identify images of moving vehicles (step S52). If multiple images of moving vehicles are captured simultaneously, each vehicle image is processed in parallel.

[0025] From the image of the moving vehicle, it is determined whether or not the vehicle is within the managed area 102 (step S53). If a vehicle falls within the managed area 102, it is considered a vehicle eligible for addition (step S54).

[0026] If the vehicle is not one that enters the managed area 102, it is determined whether or not it is a vehicle that leaves the managed area 102 (step S55). If a vehicle is leaving the managed area 102, it is treated as a vehicle subject to deduction (step S56). If it is determined that the vehicle is not leaving the managed area 102, it is treated as a vehicle that is not subject to addition or subtraction (step S57).

[0027] For vehicles subject to addition (vehicles entering the managed area 102), a value of +1 is output; for vehicles subject to subtraction (vehicles leaving the managed area 102), a value of -1 is output; and for vehicles not involved, a value of 0 is output (step S58). The aggregation means 212 then performs the aggregation. Note that the order of steps S53 and S55 can be reversed.

[0028] In this embodiment 1, an example of the output of the vehicle detection means 211 and the aggregation state by the aggregation means 212 is shown in Figure 6. In Figure 6, the vertical axis represents the number of vehicles, and the horizontal axis represents time. When a vehicle enters the managed area 102, a value of +1 is output, and when it leaves the managed area 102, a value of -1 is output. Each time a vehicle is detected, the detection result is recorded and output as shown in bar graph A at the bottom of the figure. By adding or subtracting the detection results in bar graph A using the aggregation means 212, the cumulative value is calculated as shown in step graph B.

[0029] In this embodiment 1, based on image information from the imaging device 10, the number of vehicles 51 leaving the managed area 102 and vehicles 52 entering the managed area 102 are detected, and the number of vehicles entering is added and the number of vehicles leaving is subtracted to determine the total number of vehicles present in the managed area 102. In other words, the entry and exit of vehicles are detected by detecting images of vehicles crossing the perimeter of the managed area 102 and the direction of movement of those vehicle images, and these detection results are compiled.

[0030] The aisle image information processing means 230 shown in Embodiment 1 includes a vehicle identification means 231, a parking position estimation means 232, and a monitoring means 233, as shown in Figure 1. Here, the aisle image information processing means 230 processes information on which parking space 101 a moving vehicle 53 traveling along the aisle 103 will park in, as shown in Figure 7.

[0031] The vehicle identification means 231 shown in Figure 1 recognizes moving vehicles 53 in the passage 103 based on image information from the imaging device 10, grasps the characteristics of the moving vehicles 53 from the images of the moving vehicles 53, and distinguishes each moving vehicle 53 to obtain information on its position and direction of movement.

[0032] The parking position estimation means 232 estimates which parking space 101 the moving vehicle 53 will park in, based on image information from the imaging device 10, the arrangement information of the parking spaces 101, and the position information and direction of movement information of the moving vehicle 53.

[0033] The monitoring means 233 receives outputs from the vehicle identification means 231 and the parking position estimation means 232 and monitors whether the moving vehicle 53 parks in the estimated parking space 101. If the moving vehicle 53 identified by the vehicle identification means 231 does not park in the parking space 101 estimated by the parking position estimation means 232, the monitoring means 233 instructs the parking position estimation means 232 to estimate a new parking space 101 in accordance with the movement of the moving vehicle 53.

[0034] Therefore, it is not a problem if the estimation of the parking space 101 is not always accurate, and by repeating the estimation as the moving vehicle 53 moves, it is ultimately possible to link the information of the moving vehicle 53 with the information of the parking space 101.

[0035] Furthermore, since the vehicle identification means 231 outputs the direction of movement of the moving vehicle 53 to the monitoring means 233, the monitoring means 233 can detect when the moving vehicle 53 is moving in the wrong direction and output that information to the management device 22.

[0036] The management device 22 can perform management functions such as issuing a warning to the corresponding moving vehicle 53 when it receives a signal from the monitoring means 233, such as a signal indicating wrong-way driving. Furthermore, by understanding the movement of the moving vehicle 53 using the aisle image information processing means 230, information can be created to guide the moving vehicle 53 to the parking space 101.

[0037] Embodiment 2. This second embodiment is a modification of the configuration of the first image information processing means 210 and the second image information processing means 220 of the first embodiment, but the other configurations are the same as those of the first embodiment, so their explanation will be omitted.

[0038] In other words, as shown in Figure 8, this second embodiment adds an image information analysis means 213 to the first image information processing means 210 and an image information analysis means 223 to the second image information processing means 220 as means for determining the number of vehicles in the managed area 102.

[0039] Figure 8 shows the configuration of the parking management system 1000 according to Embodiment 2. The first image information processing means 210 and its surroundings shown in Figure 8 will be described below.

[0040] The first image information processing means 210 includes a vehicle detection means 211, an aggregation means 212, and an image information analysis means 213. Furthermore, as shown in Figure 2, the first image information processing means 210 is assigned to process image information of one of the predetermined areas of the parking lot 100, which is designated as the managed area 102.

[0041] The image information analysis means 213 uses image information from the imaging device 10 to determine the shape of the vehicles 50 that appear in the image information of the managed area 102 (including the shape of the vehicle body that is partially hidden and not visible), thereby grasping information about the vehicles 50 in the parking spaces 101 of the managed area 102, and outputs the number of vehicles included in the managed area 102 to the aggregation means 212.

[0042] The vehicle detection means 211, in the same manner as shown in Embodiment 1, uses image information from the imaging device 10 to detect vehicles 51 leaving the managed area 102 and vehicles 52 entering the managed area 102, and outputs the detection results to the aggregation means 212.

[0043] The aggregation means 212 receives the output from the vehicle detection means 211 and the image information analysis means 213 and aggregates the number of vehicles 50 present in the managed area 102. The aggregation results from the aggregation means 212 and 222 of the first image information processing means 210 and the second image information processing means 220, respectively, are output to the management device 22.

[0044] The management device 22 receives the outputs of the aggregation means 212 and 222 of the first image information processing means 210 and the second image information processing means 220, respectively, and aggregates the usage status of the entire parking lot 100.

[0045] Next, the processing of the image information analysis means 213 will be explained using the managed area 102, labeled with reference numeral B in Figure 2, as an example. The imaging device 10 captures the managed area 102 of B and obtains the overhead image shown in Figure 3.

[0046] The overhead image in Figure 3 is analyzed by the image information analysis means 213 and it is determined that the managed area 102 contains multiple parking spaces 101, some of which are occupied by vehicles 50 and are in a "full" state, and some of which are in an "empty" state. Further detailed analysis reveals that the managed area 102 of B has 32 parking spaces 101, of which 24 are "full" and 8 are "empty".

[0047] Furthermore, in the analysis of the overhead image shown in Figure 3, by using deep learning inference to determine whether a vehicle has been detected and whether it is parked in a parking space 101, the managed area can be represented in a plan view as shown in Figure 4, and the "full" or "empty" state of the parking space 101 can be estimated.

[0048] In Figure 4, of the 32 parking spaces 101 in the managed area 102 of B, 24 are in use, two vehicles 51 are leaving, one vehicle 52 is about to enter, and 21 vehicles 50 are parked in the parking spaces 101.

[0049] In this second embodiment, Figure 9 shows an example of the results obtained by aggregating the output of the image information analysis means 213 using the aggregation means 212. In Figure 9, the vertical axis represents the number of vehicles, and the horizontal axis represents time. Line graph C shows the number of vehicles as determined at predetermined time intervals by the image information analysis means 213. In this example, the number of vehicles was 0 when the vehicle counting began, and changed to 3, 2, 5, 11, 18, 17, 22, and 26 as the predetermined time intervals progressed.

[0050] Figure 10 shows Figure 9 superimposed on Figure 6. Figure 10 shows that the number of vehicles identified by the image information analysis means 213 shown in Figure 9 matches the cumulative number of vehicles detected by the vehicle detection means 211 shown in Figure 6.

[0051] As shown in Figure 10, when the cumulative values ​​of the analysis results from the image information analysis means 213 and the detection results from the vehicle detection means 211 match, the changes between the periodic counts of vehicles detected by the image information analysis means 213 can be compensated for by adding or subtracting the detection results from the vehicle detection means 211 to the analysis results from the image information analysis means 213 (number of vehicles identified) in the aggregation means 212. The flowchart of this procedure is shown in Figure 11.

[0052] Figure 11 shows the aggregation flow by the aggregation means 212. Specifically, aggregation starts from a predetermined point in time (step S110). Then, the image information analysis means 213 determines the number of vehicles (step S111). In addition, the vehicle detection means 211 detects vehicles entering and leaving the managed area 102 (step S112).

[0053] In the aggregation means 212, the number of vehicles detected by the vehicle detection means 211 is added to or subtracted from the number of vehicles grasped by the image information analysis means 213 (step S113). This makes it possible to grasp the change in the number of vehicles from a predetermined point in time.

[0054] The first image information processing means 210 determines the number of vehicles parked in a predetermined management area 102 within the parking lot 100, and also detects the number of vehicles entering and exiting the management area 102, thereby outputting the usage status of the management area 102 to the management device 22.

[0055] The second image information processing means 220 is assigned a different managed area 102 than the first image information processing means 210, and operates in the same manner as the first image information processing means 210, outputting the usage status of the managed area 102 to the management device 22.

[0056] In Embodiment 2, it was explained that the image information analysis means 213 periodically determines the number of vehicles, and that the vehicle detection means 211 detects vehicles within this period. However, the period for determining the number of vehicles by the image information analysis means 213 can be shortened.

[0057] For example, as shown in Figure 12, by determining the number of vehicles from the image information using the image information analysis means 213 at intervals of a few seconds, the time it takes for vehicles 51 and 52 to move in and out of the managed area 102 shown in Figure 6 can be shortened.

[0058] Figure 12, similar to Figure 10, shows graph B, which aggregates the detection results by the vehicle detection means 211, with the vertical axis representing the number of vehicles and the horizontal axis representing time. Graph C shows the results of determining the number of vehicles by the image information analysis means 213. Graph C represents an example where the period for determining the number of vehicles by the image information analysis means 213 is set to 1 second.

[0059] In other words, since the output from the vehicle detection means 211 is determined by the time required for vehicles to enter and exit, by making the cycle for determining the number of vehicles by the image information analysis means 213 shorter than the detection output of the vehicle detection means 211, it is possible to shorten the information processing time for determining the number of vehicles.

[0060] Embodiment 3. This third embodiment describes the errors that are anticipated in the second embodiment and the countermeasures to address them. The error that occurs when determining the number of vehicles in the managed area 102 using the image information analysis means 213 is related to the number of vehicles in the managed area 102. In other words, as the number of vehicles increases, vehicles behind them are hidden by the shadows of other vehicles, resulting in more overlap in the vehicle images.

[0061] Furthermore, the error that occurs when the results of vehicle detection by the vehicle detection means 211 are aggregated to determine the number of vehicles present in the managed area 102 is due to the overlapping of images of vehicles moving simultaneously behind the vehicle with the image of the vehicle entering or leaving.

[0062] Figure 13 shows the relationship between the number of vehicles in the managed area 102 and the error. In the figure, the horizontal axis represents the number of vehicles, and the vertical axis represents the error. In the figure, the line labeled a shows the error state by the image information analysis means 213, and the line labeled b shows the error state of the aggregation of detection results by the vehicle detection means 211.

[0063] As shown in Figure 13, the error by the image information analysis means 213 increases as the number of vehicles in the managed area 102 increases. However, the error by the vehicle detection means 211 is based on vehicles entering and exiting, so the error occurs in a state that remains almost constant at all times, and is therefore represented as always containing some error, as shown by the line labeled b.

[0064] Therefore, as the number of vehicles in the managed area 102 increases, aggregating the detection results of the vehicle detection means 211 can keep the error smaller than determining the number of vehicles using the image information analysis means 213. For this reason, the number of vehicles at which the line labeled a and the line labeled b intersect is set as a first threshold (labeled X1), and when the number of vehicles is smaller than the first threshold, the image information analysis means 213 is used, and when the number of vehicles is larger than the first threshold, the vehicle detection means 211 is used. By switching to this method, the increase in error can be suppressed.

[0065] Figure 14 shows the processing flow of the image information analysis means 213, vehicle detection means 211, and aggregation means 212 in this third embodiment.

[0066] First, when starting the flow in Figure 14 (step S140), a first threshold is set for the number of vehicles that can be accurately determined using image information. For example, if it is considered that the number of vehicles can be accurately determined based on image information if there are 20 vehicles or fewer, the first threshold is set to 20 vehicles.

[0067] Then, the number of vehicles in the managed area 102 of the parking lot 100 is determined by the image information analysis means 213 (step S141). It is determined whether the determined number of vehicles is below a first threshold (step S142).

[0068] If the number of identified vehicles is below the first threshold, the number of vehicles determined by the image information analysis means 213 is considered correct, and the number of vehicles is continued to be determined by the image information analysis means 213 (step S143).

[0069] If the number of vehicles identified by the image information analysis means 213 exceeds the first threshold, the number of vehicles detected by the vehicle detection means 211 is added to or subtracted from the previously aggregated number of vehicles by the aggregation means 212 to determine the number of vehicles present in the managed area 102 (step S144).

[0070] In this embodiment 3, the error caused by an increase in the number of vehicles in the managed area 102 is prevented by setting the number of vehicles that is expected to cause the error as a first threshold, and when it is determined that the number of vehicles exceeds the first threshold, the vehicle identification by the image information analysis means 213 is switched to aggregating the number of vehicles detected by the vehicle detection means 211, thereby preventing an increase in the error in the number of vehicles.

[0071] Embodiment 4. In this fourth embodiment, we will describe the errors that can be expected when aggregating the detection results of the vehicle detection means 211 and countermeasures to address them.

[0072] The error that occurs when aggregating the results of vehicle detection by the vehicle detection means 211 to determine the number of vehicles in the managed area 102 is due to the overlapping of images of vehicles moving simultaneously behind the images of vehicles entering or leaving the area. Furthermore, this error occurs during vehicle detection by the vehicle detection means 211, and this error is accumulated in the cumulative total calculated by the aggregation means 212.

[0073] In other words, if vehicle detection by the vehicle detection means 211 is continued, errors will accumulate depending on the length of time it is used.

[0074] Figure 15 shows the relationship between the number of vehicles in the managed area 102 and the error. Similar to Figure 13, in Figure 15, the line labeled a indicates the error state of the image information analysis means 213, and the line labeled b indicates the error state of the aggregation of detection results by the vehicle detection means 211.

[0075] Here, we assume a scenario where the number of vehicles in the managed area 102 decreases. When the number of vehicles decreases, the error becomes larger when using the vehicle detection means 211 than when using the image information analysis means 213. Therefore, the number of vehicles at which the line labeled a and the line labeled b intersect is set as a second threshold (labeled X2), and when the number of vehicles falls below the second threshold, the increase in error can be suppressed by switching from the use of the vehicle detection means 211 to the use of the image information analysis means 213.

[0076] Even parking lots that are generally open 24 hours a day have periods when the number of vehicles decreases. For example, late at night or early in the morning. In this fourth embodiment, if the number of vehicles decreases, the image information analysis means 213 is used to reset the number of vehicles to the correct number.

[0077] Figure 16 shows the processing flow of the image information analysis means 213, vehicle detection means 211, and aggregation means 212 in this embodiment 4.

[0078] First, when starting the flow in Figure 16 (step S160), a second threshold is set for the number of vehicles that can be reliably and accurately determined using image information. For example, if it is considered that the number of vehicles can be accurately determined based on image information if there are 10 or fewer vehicles, the second threshold is set to 10 vehicles.

[0079] Then, the number of vehicles in the managed area 102 of the parking lot 100 is tallied based on the detection results of the vehicle detection means 211 (step S161). It is determined whether the tallied number of vehicles is below a second threshold (step S162).

[0080] If the number of identified vehicles is below the second threshold, the number of vehicles determined by the image information analysis means 213 is considered correct, the cumulative number of vehicles determined by the vehicle detection means 211 up to that point is discarded, and the number of vehicles determined by the image information analysis means 213 is used (step S163).

[0081] If the aggregated value obtained by adding or subtracting the number of vehicles detected by the vehicle detection means 211 exceeds the second threshold, the number of vehicles determined by the image information analysis means 213 is deemed inaccurate, and the number of vehicles detected by the vehicle detection means 211 is added or subtracted from the previously aggregated number of vehicles to determine the number of vehicles present in the managed area 102 (step S164).

[0082] Figure 17 shows the relationship between the first threshold and the second threshold when using the image information analysis means 213 and when using the vehicle detection means 211. In the figure, the vertical axis represents the number of vehicles and the horizontal axis represents time. Line graph C shows the case when using the image information analysis means 213, and line graph B shows the case when using the vehicle detection means 211.

[0083] In this embodiment 4, assuming that errors will accumulate if the cumulative state continues to be used for a long period of time, when the total number of vehicles collected falls below a predetermined second threshold, the use of the vehicle detection means 211 is switched to the use of the image information analysis means 213, the accumulated value that had already been collected is discarded, and the number of vehicles is determined by the image information analysis means 213, thereby preventing an increase in errors in the number of vehicles.

[0084] Embodiment 5. This embodiment 5 proposes setting a "second threshold" for suppressing errors in multiple stages.

[0085] As explained in Embodiment 4, the "second threshold" is set because, as the number of vehicles in the managed area 102 decreases, the error becomes larger when counting the number of vehicles using the vehicle detection means 211 than when counting the number of vehicles using the image information analysis means 213. Therefore, when the number of vehicles falls below the second threshold, the system switches from using the vehicle detection means 211 to using the image information analysis means 213. However, when counting the number of vehicles using the vehicle detection means 211, the error increases over time, resulting in a large cumulative value including the error. Consequently, the cumulative value deviates from the second threshold.

[0086] Therefore, in this embodiment 5, the number of vehicles at the second threshold is varied according to the elapsed time.

[0087] Figure 18 shows the relationship between the second threshold, which is set according to the passage of time, and the error. In the figure, the horizontal axis represents time, and the vertical axis represents the error. In the figure, the line labeled c shows the state of the error by the vehicle detection means 211, and it shows that when the number of vehicles is tallied using the vehicle detection means 211, the error increases almost in direct proportion to time. As shown in the figure, by varying the second threshold value as X2-1, X2-2, X2-3, X2-4, X2-5 over time, the threshold value becomes one that corresponds to the accumulation of errors over time.

[0088] By varying the second threshold over time, the transition from using the vehicle detection means 211 to using the image information analysis means 213 is made easier. This prevents the vehicle count detection means 211 from continuing to perform calculations over a long period of time, thereby preventing the error from continuously increasing.

[0089] Furthermore, to prevent an increase in errors, Image information analysis means 213 From use Vehicle detection means 211 By switching to the image information analysis means 213 according to the elapsed time since the switch occurred, an increase in errors can be prevented. Also, Image information analysis means 213 By scheduling the switch to be executed at a predetermined time, the increase in errors can be prevented.

[0090] In all embodiments 1 to 5, the information processing system 20 consists of a processor 130 and a storage device 131, as shown in Figure 19 as an example of the hardware. Although the configuration of the storage device 131 is not shown, it includes a volatile storage device such as random access memory and a non-volatile auxiliary storage device such as flash memory. Alternatively, a hard disk may be provided as an auxiliary storage device instead of flash memory. In this case, a program is input from the auxiliary storage device to the processor 130 via the volatile storage device, and the processor 130 executes this program. The processor 130 may also output data such as calculation results to the volatile storage device of the storage device 131, or it may store the data in the auxiliary storage device via the volatile storage device.

[0091] Although this application describes various exemplary embodiments and examples, the various features, aspects, and functions described in one or more embodiments are not limited to the application of a particular embodiment, but can be applied individually or in various combinations to the embodiments. Accordingly, countless variations not illustrated are conceivable within the scope of the technology disclosed herein. These include, for example, modifications, additions, or omissions of at least one component, as well as the extraction of at least one component and its combination with components of other embodiments. [Explanation of symbols]

[0092] 10 Imaging device, 20 Information processing system, 21 Image information processing device, 22 Management device, 50, 51, 52 Vehicle, 53 Mobile vehicle, 100 Parking lot, 101 Parking space, 102 Managed area, 103 Aisle, 130 Processor, 131 Storage device, 210 First image information processing means, 211, 221 Vehicle detection means, 212, 222 Aggregation means, 213, 223 Image information analysis means, 220 Second image information processing means, 230 Aisle image information processing means, 231 Vehicle identification means, 232 Parking position estimation means, 233 Monitoring means, 1000 Parking lot management system

Claims

1. A parking management system comprising an imaging device that photographs a parking lot and outputs image information, and an image information processing device, The image information processing device comprises a vehicle detection means that detects vehicles entering and leaving the managed area of ​​the parking lot based on image information of the managed area of ​​the parking lot obtained by the imaging device and outputs the detection results, and an aggregation means that receives the output of the vehicle detection means and aggregates the number of vehicles present in the managed area by adding the number of vehicles entering the managed area and subtracting the number of vehicles leaving the managed area. The image information processing means includes an image information analysis means that determines the number of vehicles present in the managed area based on the image captured by the imaging device. The determination of the number of vehicles by the image information analysis means is performed periodically at predetermined time intervals based on images captured by the imaging device, and the detection by the vehicle detection means is performed each time a vehicle enters or leaves the premises. A parking management system characterized in that the period for determining the number of vehicles by the image information analysis means is set to be shorter than the period for the vehicle detection means to output the detection result.

2. A parking management system comprising an imaging device that photographs a parking lot and outputs image information, and an image information processing device, The image information processing device comprises a vehicle detection means that detects vehicles entering and leaving the managed area of ​​the parking lot based on image information of the managed area of ​​the parking lot obtained by the imaging device and outputs the detection results, and an aggregation means that receives the output of the vehicle detection means and aggregates the number of vehicles present in the managed area by adding the number of vehicles entering the managed area and subtracting the number of vehicles leaving the managed area. The image information processing means includes an image information analysis means that determines the number of vehicles present in the managed area based on the image captured by the imaging device. A parking management system characterized in that, in determining the number of vehicles by the image information analysis means, the number of vehicles that can be determined without error is set as a first threshold, and if the determined number of vehicles exceeds the first threshold, the number of vehicles detected by the vehicle detection means is aggregated by the aggregation means to determine the number of vehicles present in the managed area.

3. A parking management system comprising an imaging device that photographs a parking lot and outputs image information, and an image information processing device, The image information processing device comprises a vehicle detection means that detects vehicles entering and leaving the managed area of ​​the parking lot based on image information of the managed area of ​​the parking lot obtained by the imaging device and outputs the detection results, and an aggregation means that receives the output of the vehicle detection means and aggregates the number of vehicles present in the managed area by adding the number of vehicles entering the managed area and subtracting the number of vehicles leaving the managed area. The image information processing means includes an image information analysis means that determines the number of vehicles present in the managed area based on the image captured by the imaging device. A parking management system characterized in that the detection by the vehicle detection means is switched to determining the number of vehicles by the image information analysis means according to the elapsed time since the start of detection by the vehicle detection means, or at a predetermined time.

4. A parking management system comprising an imaging device that photographs a parking lot and outputs image information, and an image information processing device, The image information processing device comprises a vehicle detection means that detects vehicles entering and leaving the managed area of ​​the parking lot based on image information of the managed area of ​​the parking lot obtained by the imaging device and outputs the detection results, and an aggregation means that receives the output of the vehicle detection means and aggregates the number of vehicles present in the managed area by adding the number of vehicles entering the managed area and subtracting the number of vehicles leaving the managed area. The image information processing means includes an image information analysis means that determines the number of vehicles present in the managed area based on the image captured by the imaging device. A parking management system characterized in that, in determining the number of vehicles by the image information analysis means, the number of vehicles that can be determined without error is set as a second threshold, and if the determined number of vehicles is less than or equal to the second threshold, the number of vehicles determined by the image information analysis means is considered correct, the addition / subtraction result by the vehicle detection means is discarded, and the number of vehicles determined by the image information analysis means is taken as the number of vehicles present in the managed area.

5. The parking management system according to claim 4, characterized in that the second threshold is varied so as time passes that it increases by a predetermined number of vehicles.

6. The parking management system according to any one of claims 1 to 5, characterized in that the parking lot is divided into a plurality of managed areas, and the image information processing means is set according to the managed area.

7. The parking management system according to any one of claims 1 to 5, characterized in that the image information processing device has a passage image information processing means comprising: a vehicle identification means for identifying a moving vehicle in the passage of the parking lot; a parking position estimation means for estimating which parking space in the managed area the moving vehicle will park in; and a monitoring means for receiving outputs from the vehicle identification means and the parking position estimation means and combining information between the moving vehicle and the parking space.

8. The parking management system according to claim 6, wherein the image information processing device has a passage image information processing means comprising: a vehicle identification means for identifying a moving vehicle in the passage of the parking lot; a parking position estimation means for estimating which parking space in the managed area the moving vehicle will park in; and a monitoring means for receiving outputs from the vehicle identification means and the parking position estimation means and combining information between the moving vehicle and the parking space.

Images