JP2023010459A - Method of accurately specifying position of object in video by road marking and so on and accurately calculating travel distance, travel speed, and so on of object - Google Patents

Abstract

To solve the problem that, when an object like a vehicle travels, large errors occur in travel distance, travel speed, and so on calculated from a position before travel and a position after travel of the object because video showing a three-dimensional space projected on a two-dimensional plane does not have depth to make it difficult to calculate a position of the object in a video captured by a video device and distances to reference road markings or peripheral facilities, buildings, and so on.SOLUTION: An installation height of a video device, a height of a leading end part of an object like a vehicle in a video captured by the video device, and a distance from the video device to a peripheral reference facility, building, or road marking are measured, and a distance from the leading end of the object to the peripheral reference facility, building, or road marking is calculated, and thus a travel distance, a travel speed, and so on during an elapses time can be accurately calculated when the object travels.SELECTED DRAWING: Figure 1

Description

The present invention relates to the position and position of objects such as vehicles, motorcycles, bicycles, and pedestrians in images captured by video equipment such as drive recorders, surveillance cameras, security cameras, speed control devices, video cameras, and electronic devices with cameras. By calculating the distance from the reference surrounding facilities and structures or road markings such as stop lines, pedestrian crossings, warning marks, speed indicators, center lines, lane boundaries, lane markings, and roadside lines, The present invention relates to a method of clearly specifying the positional relationship between objects and accurately calculating the moving distance and moving speed in the elapsed time when the object moves.

When analyzing the cause of a traffic accident, the circumstances of the accident, or the traffic situation, it is required to calculate the moving distance and moving speed of the target vehicle. In the prior art, when a three-dimensional landscape is projected onto a two-dimensional plane or an object in the image and the road marking are at the same horizontal position, they are also at the same horizontal position on the three-dimensional plane. I'm judging. This will be described with reference to FIGS. 2 and 3. FIG. The image 25 is converted into an image by the image analysis means 21, and the extracted image 31 is an image in which the object 32 and the road marking 33 are lined up side by side. In the extracted image 31, the object front end 34, which is the frontmost portion of the object 32, and the road marking front end 35, which is the frontmost portion of the road marking 33, are aligned side by side at the position of the comparison reference line 36. 34 and the road marking front end 35 are assumed to be at the same position. Next, an extracted image 37 is obtained by extracting an image in which the same object 32 and another road marking 38 are arranged side by side from the image by the same method. In the extracted image 37, the object front end 39, which is the frontmost portion of the object 32, and the road marking front end 3A, which is the frontmost portion of the road marking 38, are arranged side by side at the position of the comparison reference line 3B. 39 and the road marking front end 3A are assumed to be at the same position.

Elapsed time measuring means 22 measures the elapsed time from the extracted image 31 to the extracted image 37 . Normally, video is represented by a frame rate, and since the frame rate is the number of images per second, the elapsed time is calculated by counting the number of images from the extracted image 31 to the extracted image 37 and dividing by the frame rate. can do. Therefore, the elapsed time calculation formula is (elapsed time)=(the number of images from the extracted image 31 to the extracted image 37)/(the frame rate of the image 25).

The moving distance calculating means 23 calculates the distance that the object 32 has moved in the elapsed time calculated by the elapsed time measuring means 22 . Since the elapsed time calculated by the elapsed time measuring means 22 is until the object vehicle 32 crosses the comparison reference line 36 of the extracted image 31 and reaches the comparison reference line 3B of the extracted image 37, the moving distance of the object vehicle 32 is It is the distance from the comparison reference line 36 to the comparison reference line 3B. Therefore, the moving distance from the comparison reference line 36 to the comparison reference line 3B can be measured by actual measurement or from a precise drawing such as a map or sketch.

The moving speed calculating means 24 calculates the moving speed of the object 32 from the values of the elapsed time and the moving distance calculated by the elapsed time measuring means 22 and the moving distance calculating means 23 . Since the speed can be calculated by dividing the distance by the time, (moving speed)=(moving distance)/(elapsed time). As described above, the method of specifying the positional relationship between the object 32 and the road marking, the method of measuring the moving distance based on the specified position of the object 32, and the calculation of the moving speed of the object 32 based on the measured moving distance. This method does not consider three-dimensional perspective and depth at all, and is a method that causes a large error from the actual moving distance and moving speed of the object 32 .

When a three-dimensional landscape is projected onto a two-dimensional plane, there is no perspective depth, so the road marking far behind the target is photographed at the same position side by side with the target. It judges that the object projected on the plane and the road marking are at the same position without considering the depth of the plane. The positional relationship between objects and road markings in a three-dimensional landscape and the positional relationship between objects and road markings when projected onto a two-dimensional plane will be described with reference to FIG.

In FIG. 4, as an example, the three-dimensional positional relationship between vehicle B 42, which is an object, and a crosswalk 46, which is a road marking, is shown in two diagrams, a top view and a side view. When this state is photographed by the drive recorder 45 attached to the A car 41, an image such as photographed image 4F is photographed. The photographed image will be specifically described using three horizontal photographing lines 47 , 48 and 49 .

First, on the photographing line 49, the uppermost part 4C of the B car 42 is shown in the image, and the scenery behind the B car 42 is shown directly behind like wallpaper regardless of perspective. Next, in the photographing line 48, the lowest part 4A of the B vehicle 42 is captured, and if there is no obstacle blocking the view on the right side of the B vehicle 42, it is the front end of the white line of the pedestrian crossing 46 that exists behind the B vehicle 42. The A point 4E is reflected just beside the lowest part 4A of the B car 42. - 特許庁Finally, on the photographing line 47, the intermediate portion 4B of the B vehicle 42 is captured, and the point B 4D at the rear end of the white line of the pedestrian crossing 46 behind the B vehicle 42 is captured right beside the intermediate portion 4B of the B vehicle 42.

In this way, the actual three-dimensional positional relationship between the B car 42 and the pedestrian crossing 46 in FIG. Therefore, in the prior art, it was determined that the lowest point 4A of vehicle B 42 and the point A 4E at the front end of the white line of the pedestrian crossing 46 are at the same position. As a result, the distance traveled by vehicle B 42 is measured and the speed is calculated from the point A 4E at the front end of the white line of the pedestrian crossing 46 as the starting point. A big error will occur.

The present invention calculates the actual position of vehicle B 42 by calculating the distance from point A 4E at the front end of the white line of pedestrian crossing 46, which was determined to be the same position as vehicle B 42 at the bottom 4A in the prior art. can be accurately specified, and the moving distance and moving speed in the elapsed time can be accurately calculated when the B vehicle 42 moves. To provide a method for accurately calculating the moving distance and moving speed of an object.

In order to achieve this purpose, a drive recorder, a surveillance camera, a security camera, a speed control device, a video equipment which is equipped with a dimension measuring means 3 and a depth calculating means 4 and is judged to be at the same position in the conventional technology, In the images taken by video cameras and electronic devices with cameras, the part where the vehicle, motorcycle, bicycle, or pedestrian is closest to the video device and the surrounding facilities, structures, or road markings as a reference By calculating the distance to a certain stop line, pedestrian crossing, warning mark, speed indicator, center line, lane boundary line, lane marking, and roadside line, the positional relationship with the object is clearly specified, and the object is is moved, it is possible to accurately calculate the moving distance and moving speed in the elapsed time.

The dimension measuring means measures the height between the video equipment to be used and the road surface, and measures the height of the part where the object is closest to the video equipment and the road surface in the image captured by the video equipment. By measuring the distance from the surrounding facilities, structures, or road markings used as a reference for specifying the position of the object to the video device, the video device and the object are closest to the video device. It is possible to accurately specify the position of the portion approaching the road and the positional relationship of the surrounding facilities, structures, or road markings as a reference.

The depth calculation means measures the height between the video device and the road surface, the height between the part where the object is closest to the video device and the road surface, and the surrounding area as a reference. Based on the distance from the facilities, structures, or road markings to the video device, the distance from the part of the object closest to the video device to the surrounding facilities, structures, or road markings as a reference is calculated. can identify the exact position of the object.

When the object moves, the surrounding facilities, structures, or road markings measured by the dimension measuring means before the object moves and the surrounding facilities, structures, or road markings after the movement are used as the reference. Alternatively, in the distance from the road marking, the distance between the portion of the object closest to the video device before movement calculated by the depth calculation means and the distance to the surrounding facilities, structures, or road marking as a reference. By subtracting and adding the distance between the portion of the object that is closest to the video device after movement and the reference distance to the surrounding facilities, structures, or road markings, the movement distance of the object is accurately calculated. can be calculated. By dividing the moving distance by the elapsed time measured by the elapsed time measuring means 2, the moving speed of the object can be accurately calculated.

According to the present invention, as shown in FIG. 1, the image 7 is converted into an image, and from a plurality of images, an image in which the object is aligned side by side with the surrounding facilities, structures, or road markings is extracted from a plurality of images. means 1, elapsed time measuring means 2 for measuring the elapsed time between the image of the object before and after the movement extracted by the video analysis means when the object has moved; Dimension measuring means 3 for measuring the distance necessary to specify the positional relationship between surrounding facilities and structures or road markings and the video equipment that shot the image, using the distance measured by the dimension measuring means Depth calculation means 4 for accurately calculating the distance between the closest part of the object to the video equipment that captured the image and the surrounding facilities, structures, or road markings; Movement distance calculation means for calculating a distance from the object before movement to the object after movement, and the object for which the movement distance of the object calculated by the movement distance calculation means is measured by the elapsed time measurement means. By providing the moving speed calculating means 6 for calculating the moving speed of the object by dividing by the elapsed time from the time before the movement of the object until after the movement, the position of the object can be accurately identified and the movement of the object can be determined. In this case, it is possible to accurately calculate the moving distance and moving speed in the elapsed time.

FIG. 2 is a configuration diagram of a technique for accurately specifying the position of an object in an image from a road marking or the like and accurately calculating the moving distance, moving speed, and the like of the object according to the present invention; FIG. 2 is a configuration diagram of a technique for calculating the moving distance and moving speed of an object in an image according to the prior art; FIG. 10 is an explanatory diagram for explaining a conventional method for specifying the positional relationship between an object and a road marking; FIG. 10 is an explanatory diagram for explaining a problem with a conventional method for identifying a positional relationship between an object and a road marking; According to the present invention, the position of an object in a video image is accurately specified from a road marking or the like, and the object and the road marking are measured by the dimension measuring means 3 and the depth calculating means 4 of the method of accurately calculating the moving distance and moving speed of the object. is an explanatory diagram for explaining a method of identifying the positional relationship between the . According to the present invention, the position of the object in the image is accurately specified from road markings, etc., and the object is moved by the movement distance calculation means 5 and the movement speed calculation means 6 of the method of accurately calculating the movement distance and movement speed of the object. FIG. 10 is an explanatory diagram for explaining a method of calculating a moving distance and a moving speed when the moving distance and the moving speed are calculated; According to the method of the present invention, the position of the object in the image is accurately specified from road markings, etc., and the moving distance and moving speed of the object are accurately calculated. 3 is an explanatory diagram for explaining images to be extracted as before and after movement and the elapsed time measured by the elapsed time measuring means 2; FIG. FIG. 4 is an explanatory diagram for explaining an embodiment using a method of accurately specifying the position of an object in an image from a road marking or the like and accurately calculating the moving distance, moving speed, etc. of the object according to the present invention; 3 is a flow chart for explaining the operation procedure of the image analysis means 1 of the method of accurately specifying the position of an object in an image from a road marking or the like and accurately calculating the moving distance, moving speed, etc. of the object according to the present invention. 3 is a flow chart for explaining the operation procedure of the elapsed time measuring means 2 of the method of accurately specifying the position of an object in an image from a road marking or the like and accurately calculating the moving distance or moving speed of the object according to the present invention. 4 is a flow chart for explaining the operation procedure of the dimension measuring means 3 of the method for accurately specifying the position of an object in an image from a road marking or the like and accurately calculating the moving distance and moving speed of the object according to the present invention. 4 is a flow chart for explaining the operation procedure of the depth calculation means 4 of the method for accurately specifying the position of an object in an image from a road marking or the like and accurately calculating the moving distance, moving speed, etc. of the object according to the present invention. 4 is a flow chart for explaining the operation procedure of the movement distance calculation means 5 of the method of accurately specifying the position of an object in an image from a road marking or the like and accurately calculating the movement distance, movement speed, etc. of the object according to the present invention. 4 is a flow chart for explaining the operation procedure of the moving speed calculation means 6 of the method of accurately specifying the position of an object in an image from a road marking or the like and accurately calculating the moving distance and moving speed of the object according to the present invention.

1, 4, 5, 6, 7 and 20 show a method for accurately specifying the position of an object in an image from a road marking or the like according to the present invention and for accurately calculating the moving distance and moving speed of the object. 9, 10, 11, 12, 13 and 14. FIG. FIG. 1 is a configuration diagram of a method for specifying the position of an object in an image from a road marking or the like according to the present invention, FIG. 4 is an explanatory diagram for explaining the positional relationship between the object and the road marking, and FIG. 5 is dimension measuring means. 6 is an explanatory diagram for explaining the movement distance calculation means 5 and the movement speed calculation means 6; FIG. 7 is an illustration for explaining the image analysis means 1 and the elapsed time measurement means 2; Explanatory diagrams for explanation, FIG. 9 is a flow chart explaining the operation procedure of the video analysis means 1, FIG. 10 is a flow chart explaining the operation procedure of the elapsed time measurement means 2, and FIG. 12 is a flowchart for explaining the operation procedure of the depth calculation means 4, FIG. 13 is a flowchart for explaining the operation procedure for the movement distance calculation means 5, and FIG. 14 is a flowchart for explaining the operation procedure of the movement speed calculation means 6. be.

When the processing of the video analysis means 1 is started at 91 in FIG. A pre-movement point image 71 of an image in which the front end 73 of vehicle B 72 and the front end 75 of the pedestrian crossing are aligned side by side before movement of vehicle B 72 in FIG. Similarly, after the movement of the B vehicle 72, the post-movement point image 76 of the image in which the front end 77 of the B vehicle 72 and the stop line front end 79 are aligned side by side is extracted. At 96, the before-movement point image 71 and after-movement point image 76 are output, and at 97, the process ends.

When the processing of the elapsed time measuring means 2 is started at 101 in FIG. 10, the number of images from the pre-movement point image 71 to the post-movement point image 76 is counted at 102, and the number of images is input to the video analysis means 1 in FIG. 9 at 103. Elapsed time is calculated by dividing by the frame rate of video 98 . Normally, video is represented by a frame rate that indicates the number of images per second, so the elapsed time is (elapsed time) = (the number of images from the point image 71 before movement to the point image 76 after movement) ÷ (image 98 frame rate). The elapsed time calculated at 104 is output to the moving speed calculating means 6, and the process ends at 105. FIG.

The following five distances measured by the dimension measuring means 3 are measured using a measuring instrument such as a tape measure, or are measured and calculated on a precise drawing such as a map or sketch. When the dimension measuring means 3 starts processing at 111 in FIG. 11, at 112, the lowest height 58 of the B wheel 52 in FIG. is the same as the distance from the position to the road surface. Measure the height vertically from the ground to the road surface. Next, at 114, the distance 5B on the road surface from point A 57, which is the reference point before movement in FIG. 5, to the drive recorder 54 is measured. A distance 6H on the road surface to the recorder 62 is measured. Finally, from point A 65 at the same position as the crosswalk front end 6F in FIG. 6 from which the before-movement point image 71 in FIG. Measure the distance 68 to the point C 66 at the same position as , and output the measurement data of the five points measured at 117 to the depth calculation means 4 and the moving distance calculation means 5 , and the process ends at 118 .

In FIG. 5, the right-angled triangles Iroha and Ii-ho formed with the photographing line 56 as the oblique side satisfy the similarity condition because the angles of the triangles are all equal, and the right-angled triangles Iroha and Ini-ho are similar. When the depth calculation means 4 is started at 121 in FIG. 4 values of the measurement data of the distance 5B on the road surface to the point C and the distance 6H on the road surface from the point C 66 to the drive recorder 62 in FIG. Since the right-angled triangle A.I.H. and the right-angled triangle I.I.H. distance 5A): (distance 5B on road surface from point A 57 to drive recorder 54). From this relational expression, the distance 5A from the point A 57 in FIG. 5 to the front end 5C of the vehicle B 52 in FIG. height 58) x (distance 5B on road surface between point A 57 and drive recorder 54)/(height 59 of drive recorder 54). Similarly, since the right triangle A.I.H. and the right triangle I.I.H. are similar, the distance 69 from the point C 66 in FIG. Distance 69) = (height 58 of the bottom of vehicle B 61) x (distance 6H on road surface from point C 66 to drive recorder 62)/(height 59 of drive recorder 62). At 125, the binary value of the measurement data is output to the movement distance calculating means 5, and at 126 the process ends.

13, the distance 68 from point A 65 to point C 66 in FIG. The distance 67 from point 65 to the front end 6B of vehicle B 61 and the distance 69 from point C 66 to the front end 6C of vehicle B 61 are input. 6A) = (distance 68 from point A 65 to point C 66) - (distance 67 from point A 65 to front end 6B of vehicle B 61) + (distance 69 from point C 66 to front end 6C of vehicle B 61) Calculated by The moving distance 6A of vehicle B 61 calculated at 135 is output to the moving speed calculating means 6, and the process ends at 136. FIG.

When the moving speed calculating means 6 starts at 141 in FIG. 14, the elapsed time is input from the elapsed time measuring means 2 at 142 and the moving distance is input from the moving distance calculating means 5 at 143 . Since the speed can be obtained by dividing the distance by the time, the moving speed of the B vehicle 61 is calculated at 144 by (moving speed of the B vehicle 61)=(moving distance 6A)÷(elapsed time). At 145, the calculated moving speed of vehicle B 61 is output, and at 146, the processing ends.

According to the prior art, when vehicle B 61 moves from pre-movement point 6B to post-movement point 6C, pre-movement point image 71 in FIG. Since the position of the B vehicle 61 is determined from the image 76, the distance 68 from the A point 65 to the C point 66 is set as the moving distance. There is a large difference from the actual moving distance 6A of vehicle B 61, which is calculated by the method of accurately specifying the position of the object in the video image from the road markings, etc., and accurately calculating the moving distance, moving speed, etc. of the object according to the present invention. is occurring. The actual moving speed of car B 61 calculated from this moving distance will have a larger error.

As a result, according to the method of the present invention, the position of the object in the image is accurately specified from the road markings, etc., and the moving distance and moving speed of the object are accurately calculated. Since the distance 67 from the point A 65 to the front end 6J of the vehicle B 61 and the distance 69 from the point C 66 to the front end 6I of the vehicle B 61 can be measured without error by the dimension measuring means 3 and the depth calculating means 4, the distance of the vehicle B 61 It is clear that the moving distance, moving speed, etc. can be calculated accurately.

An embodiment of a method for accurately specifying the position of an object in an image from a road marking or the like and accurately calculating the moving distance, moving speed, etc. of the object according to the present invention will be described with reference to FIG. In many cases, the A car 8A actually moves, but it is assumed that the A car 8A is stopped for easy understanding of the embodiment. The image taken by the drive recorder 89 attached to the A vehicle 8A is imaged by the image analysis means 1, and from the image, the A point 86 like the pre-movement point image 81 and the B vehicle front end 85 are located side by side at the same position. and an image in which the point C 88 and the front end 87 of the vehicle B are in the same position side by side, such as the post-movement point image 82, are extracted.

Next, the elapsed time measuring means 2 counts the number of images from the pre-movement point image 81 to the post-movement point image 82 and divides the number by the frame rate of the image 7 captured by the drive recorder 89 to calculate the elapsed time. can. Therefore, the elapsed time is calculated by (elapsed time)=(number of images from pre-movement point image 81 to post-movement point image 82)/(frame rate of image 7).

By the dimension measuring means 3, the height 8F of the lowest part of the car B, the height 8G of the drive recorder 89, the distance 8B on the road surface from the point A 86 to the drive recorder 89, and the distance 8B on the road surface from the point C 88 to the drive recorder 89 The distance 8C is measured using a measuring instrument such as a tape measure, or is measured and calculated on a precise drawing such as a map or sketch.

In FIG. 5, the right-angled triangles A.I.C. and R.I.I.H. formed with the photographing line 56 as the oblique side satisfy the similarity condition because the angles of the triangles are all equal. For this reason, (height 8F at the bottom of vehicle B): (height 8G of drive recorder 89) = (distance 8D from point A 86 to front end 85 of vehicle B): (distance from point A 86 to drive recorder 89 The relational expression of the distance 8B) on the road surface holds. Therefore, the distance 8D from the front end 85 of the B vehicle to the point A 86 is: (Distance 8D from the front end 86 of the B vehicle to the front end 85 of the B vehicle) = (height of the lowest part of the B vehicle 8F) x (from the point A 86 to the drive recorder 89 It can be calculated by dividing the distance on the road surface to 8B)/(the height of the drive recorder 89 8G). Similarly, at point 84 after movement of vehicle B, (height 8F at the bottom of vehicle B): (height 8G of drive recorder 89) = (distance 8E from point C 88 to front end 87 of vehicle B): (point C A relational expression of distance 8C) on the road surface from 88 to drive recorder 89 holds. Therefore, the distance 8E from the point C 88 to the front end 87 of the B vehicle is (Distance 8E from the point C 88 to the front end 87 of the B vehicle)=(height of the lowest part of the B vehicle 8F)×(from the point C 88 to the drive recorder 89 It can be calculated by dividing the road surface distance 8C)/(the height of the drive recorder 89 8G).

The moving distance when moving from the point 83 before the movement of the vehicle B to the point 84 after the movement of the vehicle B calculated by the moving distance calculating means 5 is (Actual moving distance 8I of the vehicle B)=(From the point A 86 to the point C 88 distance 8H)-(distance 8D from point A 86 to front end 85 of vehicle B)+(distance 8E from point C 88 to front end 87 of vehicle B). Then, the moving speed calculation means 6 calculates the moving speed based on the elapsed time measured by the elapsed time measuring means 2 and the actual moving distance 8I of vehicle B (moving speed)=(actual moving distance 8I of vehicle B)/(elapsed time). time).

According to the prior art, the distance traveled from point 83 before movement of vehicle B to point 84 after movement of vehicle B is the same as the distance 8H from point A 86 to point C 88 and the distance 8J traveled by vehicle B calculated in the prior art. The actual movement distance 8I of Car B, which is the movement distance calculated by the method of accurately specifying the position of the object in the image from the road markings etc. according to this invention and accurately calculating the movement distance and movement speed of the object. There is a big difference between A larger error occurs in the moving speed calculated by this difference in moving distance. Therefore, by using the method of the present invention for accurately specifying the position of the object in the image from the road markings and the like and accurately calculating the moving distance and moving speed of the object, the point 83 before the movement of vehicle B and the location of vehicle B can be determined. It is clear that the position of the post-movement point 84 can be accurately specified and the movement distance, movement speed, etc. can be calculated accurately.

In the above description, the case where a drive recorder installed in a vehicle is used as a video device for acquiring the image 7 is explained, but in place of the drive recorder, a surveillance camera installed on a street corner, on a road, etc. is used. , a security camera, a speeding device, or an electronic device with a camera such as a portable video camera or a smartphone with a camera may be used. . In the above explanation, the crosswalk mark on the road surface was used as a reference for calculating the moving distance and moving speed of the vehicle. Road markings such as lane boundary lines, lane markings, and roadside lines may be used as references, or facilities and structures around the road may be used as references. Furthermore, in the above description, the case of calculating the moving distance and moving speed of a vehicle has been described, but the present invention is not limited to this, and the moving distance and moving speed of a two-wheeled vehicle, a bicycle, and a pedestrian can also be calculated. It can be implemented in the same way. In short, it can be widely applied to objects that move on the ground.

In the above description, the video analysis means 1, the elapsed time measurement means 2, the depth calculation means 4, the movement distance calculation means 5, the movement speed calculation means 6, etc. are configured using a computer device such as a personal computer. It can be realized by controlling the operation by a program according to the flowcharts of FIGS. 10 and 12-14. In this case, the pre-movement image and the post-movement image extracted by the image analysis means 1, the elapsed time calculated by the elapsed time measurement means 2, the distance calculated by the depth calculation means 4, and the movement calculated by the movement distance calculation means 5 Data such as the distance and the moving speed calculated by the moving speed calculating means 6 are appropriately stored in a storage device constituting a computer device and used. These data are also displayed on a display device. As the dimension measuring means 3, a measuring instrument for measuring the actual length such as a tape measure or a device capable of measuring the distance on a precise drawing such as a map or a sketch can be used. is input to the computer device and used in the depth calculation means 4 (FIG. 12).

According to the present invention, the method of accurately specifying the position of an object in an image from a road marking or the like and accurately calculating the moving distance or moving speed of an object can be achieved by using an image taken by a video device such as a drive recorder to show the movement of a vehicle or the like. Since it is possible to accurately calculate the position, moving distance, moving speed, etc. of an object, it can be used mainly when analyzing the cause of a traffic accident, the accident situation, or the traffic situation.

1 video analysis means 2 elapsed time measurement means 3 dimension measurement means 4 depth calculation means 5 movement distance calculation means 6 movement speed calculation means 7 image

Claims (5)

To identify the position of vehicles, motorcycles, bicycles, and pedestrians in images taken by video equipment such as drive recorders, surveillance cameras, security cameras, speed control devices, video cameras, and electronic devices with cameras. Equipped with a dimension measuring means and a depth calculating means, and is based on surrounding facilities, structures, etc., or road markings such as stop lines, pedestrian crossings, advance notice marks, speed indicators, center lines, lane boundaries, lane markings, and roadside lines. Accurately specifying the position of an object in an image from road markings, etc., and moving distance and movement of the object, characterized by clearly specifying the positional relationship with the object by calculating the distance to the object A method for accurately calculating speed and other factors. The dimension measuring means measures the position of the object, the positional relationship between the object and the surrounding facilities and structures, or the road marking, which is necessary for accurately specifying the object, the video equipment, and the peripheral equipment. 2. The method of claim 1, wherein distances and dimensions of facilities, structures, or road markings are measured. method to calculate The depth calculation means calculates the distances and dimensions of the object, the video equipment, the surrounding facilities, constructions, or road markings from the distances and dimensions of the object and the surrounding facilities, constructions, or road markings measured by the dimension measuring means. 2. A method according to claim 1, wherein the position of an object in an image is accurately specified from a road marking or the like, and the moving distance and moving speed of the object are accurately calculated. When the object moves, the moving distance calculating means calculates the object measured by the dimension measuring means before and after the movement, the video equipment, the surrounding facilities and structures, or the road marking. 2. The position of the object in the image according to claim 1 is accurately specified from road markings, etc., and the moving distance and speed of the object are accurately determined. method to calculate When the object moves, the movement speed calculation means calculates the movement distance of the object calculated by the movement distance calculation means and the elapsed time from before the movement of the object to after the movement measured by the elapsed time measurement means. 2. A method for accurately specifying the position of the object in the image from road markings and the like and accurately calculating the moving distance and moving speed of the object according to claim 1, wherein the moving speed of the object is calculated from .

Images