Vehicle speed measuring device

Abstract

To calculate a speed of a vehicle by simple processing.SOLUTION: A vehicle speed measuring instrument detects the tip position of a vehicle in a first area of a frame of a video obtained by photographing a driveway, detects the tip position of the vehicle in a second area located on a downstream side of the first area of a frame after the frame in which the tip position is detected, and calculates a speed of the vehicle on the basis of the tip position detected in the first area and the tip position detected in the second area.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a vehicle speed measurement device for measuring the speed of a vehicle. [Background technology]

[0002] There is a technology that performs image recognition processing on images captured by cameras installed around the roadway and calculates the speed of vehicles traveling on the roadway. For example, Patent Document 1 discloses a technology that recognizes the average speed of each vehicle on the roadway by performing image recognition processing on images captured by a roadside camera. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-194461 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, image recognition processing involves complex processing for vehicle recognition, and is a process that imposes a high computational load.

[0005] Therefore, an object of the present invention is to calculate the speed of a vehicle by simple processing. [Means for solving the problem]

[0006] In order to solve the above problem, the vehicle speed measurement device of the present invention includes a first leading edge position detection unit that detects the leading edge position of a vehicle in a first region of a frame of an image of a roadway, and a frame after the frame in which the leading edge position is detected. No. 1 In a second area located downstream of the first area, the vehicle has a second tip position detection unit that detects the tip position of the vehicle, and a speed calculation unit that calculates the speed of the vehicle based on the tip position detected in the first area and the tip position detected in the second area.

[0007] The vehicle speed measuring device may further include a horizontal edge detection unit that detects horizontal edges, which are horizontal edge components of the vehicle, in each frame of the video, and the first leading edge position detection unit and the second leading edge position detection unit may detect the leading edge position of the vehicle using the horizontal edges.

[0008] The vehicle speed measuring device may further include a rear end position detection unit that detects the rear end position of the vehicle in a frame after the frame in which the front end position of the vehicle is detected in the second region, and the first front end position detection unit may detect the front end position of a vehicle following the vehicle in the first region in a frame after the frame in which the rear end position is detected.

[0009] The first leading edge position detection unit may detect the leading edge position of the next vehicle in a region between the rear edge position and the rear edge of the first region.

[0010] The second leading edge position detection unit may detect the leading edge position of the next vehicle in a region between the rear end position and the rear end of the second region in a frame after the frame in which the leading edge position of the next vehicle is detected by the first leading edge position detection unit. [Effects of the Invention]

[0011] According to the present invention, it is possible to calculate the speed of a vehicle through simple processing. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a diagram showing a vehicle speed measurement device 100 according to an embodiment of the present invention. [Figure 2] Photographing with the camera 110 will now be described. [Figure 3] FIG. 10 is a diagram illustrating a horizontal edge. [Figure 4] FIG. 10 is a diagram illustrating an example of horizontal edge detection. [Figure 5]FIG. 2 is a diagram illustrating a first area A1 and a second area A2. [Figure 6] 10A and 10B are diagrams illustrating detection of the leading edge position and the trailing edge position of a vehicle. [Figure 7] FIG. 4 is a diagram showing the processing operations performed in the control unit 120 when measuring the speed of a vehicle. [Figure 8] FIG. 10 is a diagram showing an example of a processing operation performed in a control unit 120 when detecting the rear end of a vehicle. DETAILED DESCRIPTION OF THE INVENTION

[0013] <Vehicle speed measuring device 100> 1 is a diagram showing a vehicle speed measurement device 100 according to an embodiment of the present invention. The vehicle speed measurement device 100 includes a camera 110 and a control unit 120.

[0014] The camera 110 photographs the roadway RW. For example, as shown in FIG. 2(A), the camera 110 photographs the roadway RW from above. Therefore, in this embodiment, an image such as that shown in FIG. 2(B) is obtained. That is, the image photographed by the camera 110 according to this embodiment captures horizontal edge components of the vehicle AM, such as the front end of the vehicle AM ​​traveling on the roadway RW and the upper and lower edges of the windshield.

[0015] The control unit 120 includes a horizontal edge detection unit 121 , a first leading edge position detection unit 122 , a second leading edge position detection unit 123 , a speed calculation unit 124 , and a trailing edge position detection unit 125 .

[0016] The horizontal edge detection unit 121 detects horizontal edges, which are horizontal edge components of a vehicle, in each frame of the video captured by the camera 110. For example, as shown in FIG. 3, the horizontal edge detection unit 121 detects horizontal edges of a vehicle in a horizontal edge detection area AE, which is a predetermined area of ​​one lane. When a vehicle AM ​​is captured in the horizontal edge detection area AE as shown in FIG. 3(A), the horizontal edge detection unit 121 detects multiple horizontal edges ED at the position where the vehicle AM ​​is captured, as shown in FIG. 3(B). Each of the multiple horizontal edges ED corresponds to the front edge of the vehicle, the upper edge or lower edge of the windshield, etc. A vehicle is present between two horizontal edges at the upper and lower ends of the multiple horizontal edges ED. The horizontal edge detection unit 121 may use any technology capable of detecting horizontal edges. Techniques for detecting horizontal edges are disclosed, for example, in Japanese Patent No. 3516118 and Japanese Patent Laid-Open No. 2002-32747.

[0017] The horizontal edge detection unit 121 may detect positions where horizontal edge components exist as horizontal edges. The horizontal edge detection unit 121 may, for example, calculate the average brightness (i.e., the sum of the brightnesses of the pixels on that line divided by the number of pixels on that line) for each line in the horizontal edge detection area AE in the direction perpendicular to the roadway traveling direction TD (horizontal direction). When a vehicle is present on the roadway as shown in FIG. 4(A), plotting the average brightness of each roadway line results in a graph like the one shown in FIG. 4(B). As shown in FIG. 4(B), the average brightness is high in areas where the vehicle's horizontal edge components exist. As shown in FIG. 4(B), by appropriately setting a threshold, lines with an average brightness above the threshold can be identified as lines where the vehicle's horizontal edge components exist. Therefore, the horizontal edge component unit 121 may detect the positions of lines with an average brightness above the threshold as horizontal edges. When setting the threshold, it is preferable to use the average brightness when no vehicle is present on the roadway, i.e., the average brightness of the background. In this case, for example, a value that is a constant multiple of the average brightness of the background may be set as the threshold value.

[0018] The first leading edge position detector 122 detects the leading edge position of the vehicle in a first region A1 of the horizontal edge detection region AE. As shown in Fig. 5, the first region A1 is the region between two lines that are perpendicular to the traveling direction TD of the roadway within the edge detection region AE. Of these two lines, the line that is forward in the traveling direction TD is defined as the leading edge F1, and the line that is rearward in the traveling direction TD is defined as the rear edge R1.

[0019] The first leading edge position detection unit 122 first checks in the detection start frame whether a horizontal edge exists in the region of the first region A1 between the first leading edge position detection start line SL1 and the rear end R1 of the first region A1 (first leading edge position detection region S1). If no horizontal edge exists in the first leading edge position detection region S1, the first vehicle end detection unit 122 checks in the next frame whether a horizontal edge exists in the first region A1. If a horizontal edge exists in the first leading edge position detection region S1, the first vehicle end detection unit 122 detects the horizontal edge EF closest to the first leading edge position detection start line SL1 as the leading edge position of the vehicle. The leading edge position of the vehicle detected in the first region A1 is the first leading edge position. The detection start frame and the first leading edge position detection start line SL1 will be described in detail below.

[0020] 6A and 6B are diagrams illustrating the detection of the leading edge position of a vehicle. Fig. 6A shows the edge detection area AE of the detection start frame (frame A), and Fig. 6B to 6J show the edge detection areas AE of frames B to J that follow frame A.

[0021] In the example shown in Fig. 6, no horizontal edge exists in the first tip position detection region S1 of frames A and B, but a horizontal edge exists in the first tip position detection region S1 of frame C. For this reason, in the example shown in Fig. 6, the first tip position detection unit 122 confirms that no horizontal edge exists in the first tip position detection region S1 of the detection start frame (frame A), and then confirms that no horizontal edge exists in the first tip position detection region S1 of frame B. Then, after confirming that no horizontal edge exists in the first tip position detection region S1 of frame B, the first tip position detection unit 122 confirms that a horizontal edge exists in the first tip position detection region S1 of frame C, and detects the horizontal edge EF closest to the first tip position detection start line SL1 in frame C as the tip position of the vehicle (first tip position).

[0022] The second leading edge position detector 123 detects the leading edge position of the vehicle in a second region A2 of the horizontal edge detection region AE. As shown in FIG. 5, the second region A2 is the region between two lines that are perpendicular to the traveling direction TD of the roadway within the edge detection region AE. The second region A2 is located forward of the leading edge F1 of the first region in the traveling direction TD of the roadway, that is, downstream of the first region A1. Of the two lines that define this second region A2, the line that is forward in the traveling direction TD is the leading edge F2, and the line that is rearward in the traveling direction TD is the rear edge R2.

[0023] In the frame following the frame in which the first leading edge position detection unit 122 detects the vehicle's leading edge position (first leading edge position) in the first region A1, the second leading edge position detection unit 123 checks whether a horizontal edge exists in the region of the second region A2 between the second leading edge position detection start line SL2 and the rear end R2 of the second region A2 (second leading edge position detection region S2). If no horizontal edge exists in the second leading edge position detection region S2, the second leading edge position detection unit 123 checks whether a horizontal edge exists in the second region A2 in the next frame. If a horizontal edge exists in the second leading edge position detection region S2, the second leading edge position detection unit 123 detects the horizontal edge closest to the second leading edge position detection start line SL2 as the vehicle's leading edge position. The vehicle's leading edge position detected by the second leading edge position detection unit 123 is the second leading edge position. The second leading edge position detection start line SL2 will be described in detail below.

[0024] In the example shown in Fig. 6, there is no horizontal edge in the second leading edge position detection region S2 of frames D and E following frame C in which the vehicle leading edge position (first leading edge position) was detected in the first region A1, but there is a horizontal edge in the second leading edge position detection region S2 of frame F. For this reason, in the example shown in Fig. 6, the second leading edge position detection unit 123 confirms that there is no horizontal edge in the second leading edge position detection region S2 of frame D, which is the frame following frame C in which the vehicle leading edge position (first leading edge position) was detected in the first region A1, and then confirms that there is no horizontal edge in the second leading edge position detection region S2 of frame E. Then, after confirming that there is no horizontal edge in the second leading edge position detection region S2 of frame E, the second leading edge position detection unit 123 confirms that there is a horizontal edge in the second leading edge position detection region S2 of frame F, and detects the horizontal edge EF closest to the second leading edge position detection start line SL2 in frame F as the vehicle leading edge position (second leading edge position).

[0025] The speed calculation unit 124 calculates the speed of the vehicle based on the first tip position, the second tip position, and the time between the time when the first tip position is detected and the time when the second tip position is detected.

[0026] The time between the time when the first tip position is detected and the time when the second tip position is detected may be obtained by counting the time from the time when the first tip position is detected to the time when the second tip position is detected using a counter, or may be calculated using the time when the first tip position is detected and the time when the second tip position is detected.

[0027] In this way, in this embodiment, the vehicle speed is calculated only by detecting horizontal edges, which makes it possible to calculate the vehicle speed through simpler processing than image recognition processing. As a result, in this embodiment, it is possible to reduce costs and man-hours.

[0028] The rear edge detection unit 125 detects the rear edge position of the vehicle in the horizontal edge detection area AE.

[0029] In a frame (rear end detection start frame) in which the second leading end position detection unit 124 detects the leading end position of the vehicle (second leading end position) in the second region A2, the rear end detection unit 125 detects the horizontal edge closest to the rear end R1 of the first region A1 as the rear end position of the vehicle. Thereafter, the rear end position detection unit 125 also detects the rear end position of the vehicle in frames following the rear end detection start frame. At this time, in each frame following the rear end detection start frame, the rear end position detection unit 125 detects the horizontal edge closest to the rear end position detection start line SL3 as the rear end position of the vehicle, among the horizontal edges located between the line on which the rear end position of the vehicle detected in the previous frame is located (rear end position detection start line SL3) and the front end F2 of the second region A2. The rear end position detection unit 125 continues detecting the rear end position of the vehicle until no horizontal edge exists between the rear end position detection start line SL3 and the front end F2 of the second region A2.

[0030] In the example shown in FIG. 6, in frame F (rear end detection start frame), in which the front end position of the vehicle (second front end position) is detected in the second region A2, the horizontal edge ER closest to the rear end R1 of the first region A1 is detected as the rear end position of the vehicle. Thereafter, the rear end position detection unit 125 also detects the rear end position of the vehicle in frames G to J following frame F. At this time, in each of frames G to J, the rear end position detection unit 125 detects the horizontal edge ER closest to the rear end position detection start line SL3, among the horizontal edges between the line on which the rear end position of the vehicle detected in the previous frame is located (rear end position detection start line SL3) and the front end F2 of the second region A2, as the rear end position of the vehicle. For example, the rear end position detection start line SL3 in frame G is the line on which the rear end position of the vehicle detected in frame F previous to frame G is located. The rear end position detection unit 125 continues to detect the rear end position of the vehicle up to frame J, in which there is no longer a horizontal edge between the rear end position detection start line SL3 and the front end F2 of the second area A2.

[0031] The distance between the front end F1 of the first area A1 and the front end F2 of the second area A2 and the distance between the rear end R1 of the first area A1 and the front end F2 of the second area A2 should be set so that when the front end of the vehicle is located within the second area A2, the rear end of the vehicle is located within the first area A1. In this way, as shown in Figure 6(F), in the frame (frame F) in which the front end of the vehicle is detected in the second area A2, the rear end of the vehicle can be detected in the first area A1.

[0032] At this time, the rear end position detection unit 125 may set the rear end position of the vehicle detected in the first area A1 as the first leading end position detection start line SL1, and the rear end position detected in the second area A2 as the second leading end position detection start line SL2. That is, the rear end position of the vehicle detected in the first area A1 by the rear end position detection unit 125 may be set as the first leading end position detection start line SL1 when detecting the leading end position of the next vehicle, and the rear end position of the vehicle detected in the second area A2 by the rear end detection unit 125 may be set as the second leading end position detection start line SL2 when detecting the leading end position of the next vehicle. In addition, the detection start frame may be a frame after the frame in which the rear end position of the immediately preceding vehicle was detected in the first area A1.

[0033] In the example shown in FIG. 6, the rear end position of the vehicle in the first area A1 of the frame F is detection After that, frame G, which is the next frame after frame F, is set as the detection start frame, and the leading edge position of the next vehicle is detected in the first leading edge position detection area S1 of frames G to I. Then, the first leading edge position detection line SL1 in frames G to I is First area A1 is the rear end position of the vehicle detected at

[0034] 6, after the leading edge position of the next vehicle is detected in the first leading edge position detection area S1 of frame I, the leading edge position of the next vehicle is detected in the second leading edge position detection area S2 of frame J, which follows frame I. The second leading edge position detection line SL2 in frame J is the rear edge position of the vehicle detected in the second area A2 of frame I.

[0035] By doing this, even if multiple horizontal edges are detected as horizontal edge components of one vehicle, it becomes possible to detect the front end position of each vehicle from among the multiple horizontal edges detected by the horizontal edge detection unit 121.

[0036] <Processing Operation in Control Unit 120> 7 is a diagram showing an example of a processing operation performed by the control unit 120 when measuring the vehicle speed. The first tip position detection unit 122 checks whether or not a horizontal edge exists in the first tip position detection region S1 in the detection start frame (step S701).

[0037] If no horizontal edge exists in the first leading edge position detection area S1 (step S701, NO), the first vehicle end detection unit 122 checks whether or not a horizontal edge exists in the first leading edge position detection area S1 in the next frame (step S701).If no horizontal edge exists in the first leading edge position detection area S1 in the next frame either (step S701, NO), the first vehicle end detection unit 122 checks whether or not a horizontal edge exists in the first leading edge position detection area S1 in the next frame (step S701).

[0038] If a horizontal edge exists within the first tip position detection area S1 (step S701, YES), the first vehicle end detection unit 122 detects the horizontal edge EF closest to the first tip position detection start line SL1 as the tip position of the vehicle (first tip position) (step S702).

[0039] The second tip position detection unit 123 checks whether or not a horizontal edge exists in the second tip position detection region S2 in the frame following the frame in which the first tip position was detected (step S703).

[0040] If no horizontal edge exists in the second tip position detection area S2 (step S703, NO), the second wheel tip position detection unit 123 checks whether or not a horizontal edge exists in the second tip position detection area S2 in the next frame (step S703).If no horizontal edge exists in the second tip position detection area S2 in the next frame either (step S703, NO), the second tip position detection unit 123 checks whether or not a horizontal edge exists in the second tip position detection area S2 in the next frame (step S703).

[0041] If a horizontal edge exists within the second tip position detection area S2 (step S703, YES), the second tip position detection unit 123 detects the horizontal edge EF closest to the second tip position detection start line SL2 as the tip position of the vehicle (second tip position) (step S704).

[0042] The speed calculation unit 124 calculates the speed of the vehicle based on the first tip position, the second tip position, and the time between the time when the first tip position is detected and the time when the second tip position is detected (step S705).

[0043] 8 is a diagram showing an example of the processing operation performed by control unit 120 when detecting the rear end of a vehicle. In a rear end detection start frame, the horizontal edge ER closest to the rear end R1 of first area A1 is detected as the rear end position of the vehicle (step S801), the detected rear end position of the vehicle is set as a first leading end position detection start line SL1 and a rear end position detection start line SL3, the frame following the rear end detection start frame is set as the detection start frame for detecting the leading end position of the next vehicle (step S802), and in the subsequent frame, the horizontal edge ER closest to the rear end position detection start line SL3 is detected as the rear end position of the vehicle (step S803).

[0044] If the detected rear end position of the vehicle is not within the second area A2 (step S804, NO), the detected rear end position of the vehicle is set as a rear end position detection start line SL3 (step S805), and in the next frame, the horizontal edge ER closest to the rear end position detection start line SL3 is detected as the rear end position of the vehicle (step S803).If the detected rear end position of the vehicle is within the second area A2 (step S804, YES), the detected rear end position of the vehicle is set as a second front end position detection start line SL2 and a rear end position detection start line SL3 (step S806).

[0045] The present invention has been described above in terms of preferred embodiments thereof. While the present invention has been described herein with reference to specific examples, various modifications and variations can be made to these examples without departing from the spirit and scope of the present invention as set forth in the claims. [Explanation of symbols]

[0046] 100 Vehicle speed measuring device 110 Camera 120 control section 121 Horizontal edge detector 122 First tip position detection unit 123 Second tip position detection unit 124 Speed ​​calculation section 125 Rear end position detector

Claims

1. a first leading edge position detection unit that detects the leading edge position of a vehicle in a first region of a frame of an image of a roadway; a second leading edge position detection unit that detects the leading edge position of the vehicle in a second region located downstream of the first region in a frame subsequent to the frame in which the leading edge position is detected; a speed calculation unit that calculates the speed of the vehicle based on a first tip position that is the tip position of the vehicle detected in the first area, a second tip position that is the tip position of the vehicle detected in the second area, and the time between the time when the first tip position is detected and the time when the second tip position is detected.

2. a horizontal edge detection unit that detects horizontal edges that are horizontal edge components of a vehicle in each of the frames of the video; The vehicle speed measurement device according to claim 1 , wherein the first leading edge position detection unit and the second leading edge position detection unit detect the leading edge position of the vehicle using the horizontal edge.

3. The vehicle further includes a rear end position detection unit that detects the rear end position of the vehicle in a frame after the frame in which the front end position of the vehicle is detected in the second region; The information processing device according to claim 1 , wherein the first leading edge position detection unit detects the leading edge position of a vehicle following the vehicle in the first region of a frame subsequent to the frame in which the rear edge position is detected.

4. 4. The vehicle speed measurement device according to claim 3, wherein the first leading edge position detection unit detects the leading edge position of the next vehicle in a region between the rear edge position detected in the first region and the rear edge of the first region.

5. 5. The vehicle speed measurement device according to claim 3, wherein the second leading edge position detection unit detects the leading edge position of the next vehicle in a frame subsequent to the frame in which the leading edge position of the next vehicle is detected by the first leading edge position detection unit, in a region between a rear end position detected in the second region and a rear end of the second region.

Images