Vehicle warning system, vehicle warning method, and vehicle warning program

The vehicle warning system enhances detection accuracy by identifying vehicles on lane boundaries and excessive speeds, issuing warnings to prevent unsafe approaches to restricted areas.

JP7742825B2Active Publication Date: 2025-09-22NAGOYA ELECTRIC WORKS
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2022197775
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2022-12-12
Publication Date
2025-09-22
Estimated Expiration
2042-12-12

AI Technical Summary

Technical Problem

Conventional vehicle detection systems struggle with accuracy in high illuminance conditions, such as daytime, and lack effective methods to warn vehicles approaching restricted areas like construction sites.

Method used

A vehicle warning system that includes an image acquisition unit, vehicle detection unit, and warning control unit to detect vehicles traveling on lane boundary lines at a threshold speed and issue warnings for abnormal behavior or excessive speed approaching restricted areas.

Benefits of technology

Effectively warns drivers of abnormal vehicle behavior, such as staying on lane boundaries or excessive speed, ensuring safety by alerting workers in restricted areas.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007742825000001
    Figure 0007742825000001
  • Figure 0007742825000002
    Figure 0007742825000002
  • Figure 0007742825000003
    Figure 0007742825000003
Patent Text Reader

Abstract

To provide a technique for warning a vehicle to approach a regulation area.SOLUTION: A vehicle warning system includes: an image acquisition section for acquiring an image of an area on a near side of a regulation area set on a road; a vehicle detection section for detecting a vehicle traveling toward the regulation area based on the image; a warning object vehicle detection section for detecting a vehicle traveling at a speed equal to or higher than a first threshold on a boundary line of a lane set on the road as a first warning object vehicle; and a warning control section for allowing a warning section to output warning information when the first warning object vehicle is detected.SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

[0001] The present invention relates to a vehicle warning system, a vehicle warning method, and a vehicle warning program. [Background technology]

[0002] Conventionally, there is known a technique for detecting a vehicle traveling on a road based on an image of the road. For example, Patent Document 1 discloses a technique for detecting the presence of a moving object when a light source having a certain illuminance enters a closed area formed in an imaging area, and for detecting the entry and exit of the moving object into and from the closed area based on a change in the average illuminance in the closed area. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6980887 Summary of the Invention [Problem to be solved by the invention]

[0004] Conventional technologies detect the entry of a moving object into a closed area based on changes in the average illuminance of an image. Therefore, the accuracy of detecting a moving object can be reduced in situations where the illuminance of the closed area is high, such as during the daytime. Furthermore, workers are often present in restricted areas such as construction sites, and if there is a vehicle that may be entering the restricted area, it is necessary to warn the work vehicle. However, there has been a lack of effective technology for detecting such vehicles.

[0005] The present invention has been made in view of the above-mentioned problems, and has an object to provide a technique for warning a vehicle that may approach a restricted area. [Means for solving the problem]

[0006] In order to achieve the above-mentioned objective, the vehicle warning system includes an image acquisition unit that acquires an image of the area in front of a restricted area set on a road, a vehicle detection unit that detects vehicles traveling toward the restricted area based on the image, a warning target vehicle detection unit that detects the vehicle traveling on a lane boundary line set on the road at a speed equal to or greater than a first threshold as a first warning target vehicle, and a warning control unit that causes the warning unit to output warning information when the first warning target vehicle is detected.

[0007] That is, the vehicle warning system detects a vehicle traveling on a lane boundary line at a speed equal to or greater than a first threshold as a first warning target vehicle and issues a warning. On roads, vehicles should travel on lanes separated by boundary lines, and are not expected to travel continuously on the boundary lines. Therefore, if a vehicle is traveling on a boundary line, it can be assumed that the driver of the vehicle is dozing off, for example. Therefore, by detecting a vehicle traveling on a lane boundary line at a speed equal to or greater than the first threshold as a warning target, it is possible to issue a warning about an abnormal vehicle that may approach a restricted area. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating a configuration of a vehicle warning system. [Figure 2] 1 is a diagram illustrating an example of a road on which a vehicle warning system is used; [Figure 3] 10 is a flowchart of a vehicle warning process. [Figure 4] FIG. 10 is a diagram for explaining a change in a time threshold value depending on the speed of a vehicle. [Figure 5] FIG. 10 is a diagram showing an example in which a vehicle travels through a first detection range. [Figure 6] FIG. 10 is a diagram showing an example in which a vehicle travels through a first detection range. DETAILED DESCRIPTION OF THE INVENTION

[0009] Here, the embodiments of the present invention will be described in the following order. (1) Vehicle warning system configuration: (2) Vehicle warning processing: (3) Other embodiments:

[0010] (1) Vehicle warning system configuration: FIG. 1 is a diagram showing the configuration of a vehicle warning system 10 according to one embodiment of the present invention. The vehicle warning system 10 includes a control unit 20, a recording medium 30, a camera 40, and a warning unit 50. In this embodiment, the vehicle warning system 10 is installed in a work vehicle used at a construction site. FIG. 2 is a diagram showing an example of a road on which the vehicle warning system 10 is used. In the example shown in FIG. 2, it is assumed that road construction work is being carried out on the road. A restricted area Ze has been set on the road to prohibit general vehicles from entering the construction site. The inside and outside of the restricted area Ze are separated by placed restricting materials C (rubber cones in the example shown in FIG. 2).

[0011] A work vehicle (not shown) is parked within the restricted area Ze. Work vehicles are used for various purposes, and are equipped with a display unit for notifying vehicles on the road that construction is underway, for example. In this embodiment, the vehicle warning system 10 is installed on the work vehicle. Specifically, the camera 40 included in the vehicle warning system 10 is installed on the work vehicle so that the area in front of the restricting material is within its field of view. This configuration allows the vehicle warning system 10 and the camera 40 to be easily installed within the restricted area Ze.

[0012] The vehicle warning system 10 monitors the area in front of the restricted area Ze, i.e., the area behind the vehicle on the road in the direction of travel as viewed from the restricted area Ze, and detects and warns of vehicles that may enter the restricted area Ze. In this embodiment, two types of behavior are subject to warning, and a vehicle that exhibits a behavior subject to warning is detected as a vehicle subject to warning. In this embodiment, one of the two types of vehicles subject to warning is called a first vehicle subject to warning, and the other is called a second vehicle subject to warning.

[0013] A first warning target vehicle is a vehicle that behaves abnormally, such as traveling on a boundary line instead of a lane in an area just before the restricted area Ze. A lane is a strip-shaped area on a road where a vehicle should travel, and a boundary line is a solid or dashed line that separates the lanes. In this embodiment, of the boundary lines Lb1 and Lb2 of the lane L1 where the restricted area Ze exists, a vehicle that is located on the left side when facing forward in the direction of travel just before the restricted area Ze and traveling on the solid boundary line Lb1 at the edge of the width of the road may be a first warning target vehicle. Normally, a vehicle traveling on the lane L1 side would change lanes to lane L2 before reaching the restricted area Ze. However, a driver of a vehicle traveling on the boundary line Lb1 on the opposite side of lane L2 without changing lanes is likely to be unaware of the restricted area Ze.

[0014] A vehicle subject to the second warning is a vehicle that has behaved by approaching the restricted area Ze excessively at an excessive speed. The behavior of traveling in an area far from the restricted area Ze on the lane L1 in front of the restricted area Ze is not abnormal in itself. However, under normal circumstances, a vehicle would not travel in such a way as to approach the restricted area Ze excessively at an excessive speed. Therefore, such a vehicle is deemed to be a vehicle subject to the second warning.

[0015] The vehicle warning system 10 is a system for detecting and warning the first and second warning target vehicles described above. The control unit 20 is a computer equipped with a CPU, RAM, ROM, etc. The control unit 20 performs vehicle warning processing by executing various programs stored in the recording medium 30 and ROM. The recording medium 30 stores various programs and various information. The various information includes image data 30a indicating an image captured by the camera 40 and parameter information 30b indicating parameters referenced when performing the vehicle warning processing.

[0016] In this embodiment, the parameter information 30b includes information for evaluating the detection range, speed, and traveling time on the boundary line of the first vehicle to be warned. The detection range of the first vehicle to be warned is the range on the boundary line Lb1 shown in FIG. 2 above, and is a range of a predetermined distance along the boundary line Lb1 from the restricted area Ze to the near side. This predetermined distance is referred to as the first distance, and this range is referred to as the first detection range. The parameter information 30b includes a value indicating the first distance. FIG. 2 shows the first distance Db1 and the first detection range Z1. Although FIG. 2 shows the first detection range as a range wider than the boundary line Lb1, it is sufficient for the first detection range to have at least the same width as the boundary line Lb1, and it is sufficient for the first detection range to have a predetermined width approximately equivalent to that of the boundary line Lb1.

[0017] Even if a vehicle is traveling within the first detection range, if the vehicle is traveling at a low speed, there is a possibility that the driver has intentionally parked the vehicle on the shoulder of the road. Therefore, in this embodiment, a minimum speed of vehicles traveling within the first detection range that are subject to a warning is specified. This minimum speed is called a first threshold value V1. The parameter information 30b includes a value indicating the first threshold value V1.

[0018] The first threshold value may be a predetermined speed value that is excessive for traveling on the boundary line Lb1 in the area before the restricted area Ze. In other words, before the restricted area Ze, the vehicle is expected to change lanes from lane L1 to lane L2 to avoid the restricted area Ze. However, if the speed does not decrease sufficiently before the restricted area Ze, it can be assumed that some abnormality has occurred, such as the driver not noticing the restricted area Ze or falling asleep at the wheel.

[0019] Furthermore, even if a vehicle has traveled through the first detection range, if it immediately returns to lane L1 and does not travel through the first detection range again, there is a possibility that the vehicle's behavior is not abnormal. Therefore, in this embodiment, if the length of time that the vehicle has traveled through the first detection range is relatively long, the vehicle is deemed to be a vehicle subject to the first warning. The parameter information 30b defines a value for determining a time threshold that is a threshold for evaluating the length of time. In this embodiment, the value is information for determining how long the vehicle has traveled within the first detection range (i.e., on the boundary line) out of the length of time that the vehicle travels through the first detection range along the boundary line, and is a value greater than 0 and less than 1. In this embodiment, the value is assumed to be 0.5.

[0020] Furthermore, the parameter information 30b includes information for evaluating the detection range and speed of the second vehicle to be warned. As shown in FIG. 2, the detection range of the second vehicle to be warned is a predetermined distance on the lane L1 from the restricted area Ze toward the front. This predetermined distance is referred to as the second distance, and this range is referred to as the second detection range. The parameter information 30b includes a value indicating the second distance. FIG. 2 shows the second distance Db2 and the second detection range Z2. Even if a vehicle is traveling within the second detection range, if the vehicle is traveling at a low speed, the driver may intend to stop or change lanes before the restricted area Ze. Therefore, in this embodiment, a minimum speed of vehicles traveling within the second detection range that are to be warned is specified. This minimum speed is referred to as the second threshold V2. The parameter information 30b includes a value indicating the second threshold V2. In this embodiment, the first threshold V1 is smaller than the second threshold V2, and the first distance Db1 is larger than the second distance Db2, but these relationships are not limited to these.

[0021] The camera 40 is equipped with a photographing optical system and an image sensor, and is a device that captures images at a predetermined frame rate. The control unit 20 acquires the images output by the camera 40 via an interface (not shown). In this embodiment, the camera 40 is installed so that the lane L1 in front of the restricted area Ze and the area around it are included in its field of view. The warning unit 50 is a device that issues a warning to workers and other people within and around the restricted area Ze. The type of warning is not limited, but in this embodiment, the warning unit 50 is a speaker that outputs a warning sound.

[0022] The control unit 20 uses the above configuration to detect vehicles that are subject to a warning and to issue a warning. To this end, the control unit 20 executes a vehicle warning program 21. When the vehicle warning program 21 is executed, the control unit 20 functions as an image acquisition unit 21a, a vehicle detection unit 21b, a warning subject vehicle detection unit 21c, and a warning control unit 21d.

[0023] The image acquisition unit 21a has a function of acquiring images of an area in front of the restricted area Ze installed on the road. That is, the control unit 20 acquires images output from the camera 40 at a predetermined frame rate using the function of the image acquisition unit 21a and records them as image data 30a on the recording medium 30. The vehicle detection unit 21b has a function of detecting vehicles traveling toward the restricted area Ze based on the images. That is, the control unit 20 detects vehicles traveling on the road in front of the restricted area Ze based on the image data 30a using the function of the vehicle detection unit 21b. The details of this processing will be described later.

[0024] The warning target vehicle detection unit 21c has a function of detecting a vehicle traveling on the boundary line of a lane L1 set on a road at a speed equal to or greater than a first threshold value V1 as a first warning target vehicle, and detecting a vehicle traveling on the lane L1 within a second detection range of a predetermined distance in front of the restricted area Ze at a speed equal to or greater than a second threshold value V2 as a second warning target vehicle. That is, the control unit 20 determines whether a vehicle detected by the function of the vehicle detection unit 21b corresponds to either a first warning target vehicle or a second warning target vehicle. Details of this process will be described later.

[0025] The warning control unit 21d has a function of causing the warning unit to output warning information when a first warning target vehicle is detected and when a second warning target vehicle is detected. That is, when a warning target vehicle is detected, the control unit 20 controls the warning unit 50 to output a warning sound using the function of the warning control unit 21d. With the above configuration, it is possible to warn of vehicles that may approach the restricted area Ze. Therefore, workers and other people within and around the restricted area Ze can notice the vehicle that is the target of the warning and can avoid the vehicle early.

[0026] (2) Vehicle warning processing: Next, the vehicle warning process will be described. Fig. 3 is a flowchart showing the vehicle warning process. The user installs the vehicle warning system 10 on a work vehicle parked in a restricted area Ze so that the field of view of the camera 40 includes the lane L1 in front of the restricted area Ze and the area around it. The vehicle warning process starts in this state.

[0027] When the vehicle warning process is started, a first detection range and a second detection range are set (step S100). The first detection range and the second detection range may be set by various methods. Here, an example will be described in which the first detection range and the second detection range are set based on image data 30a captured by the camera 40. In this example, the control unit 20 recognizes reference objects included in the image data 30a. The reference objects are objects that are referenced to identify distances on the road, such as road regulations, lines, people, etc. These reference objects may be recognized using known pattern matching or a machine learning model.

[0028] Among the lines on the road, for example, the length and spacing of dashed lines between lanes are known. Therefore, by identifying the dashed lines in the image represented by the image data 30a and determining the length and spacing of the dashed lines, the control unit 20 can determine the distance from the near edge of the restricted area Ze, which is separated by a regulating material present at the rear end of the restricted area Ze in the direction of travel, to any point on the road. Note that if there are multiple types of dashed line lengths and dashed line spacing, the user may be able to select which type to use. Furthermore, by capturing an image with the camera 40 while a person is standing at a specific distance from the near edge of the restricted area Ze, the position at that specific distance can be associated with the position in the captured image. Various other techniques can also be used to determine the distance from the near edge of the restricted area Ze on the road in the image.

[0029] In either case, the control unit 20 can determine the distance from the near end of the restricted area Ze, which is separated by the restricting material, to any position on the road in the image captured by the camera 40. The control unit 20 determines the first detection range and the second detection range by referring to the distance determined in this manner and the parameter information 30b.

[0030] The first detection range is a range of a first distance along the boundary line Lb1 on the near side of the restricted area Ze. Therefore, based on the image data 30a, the control unit 20 detects a solid white line extending from the restricted area Ze and regards the solid white line on the left side as the boundary line Lb1 when the vehicle travels forward (see FIG. 2 ; the same applies below). The control unit 20 also detects a restriction material on the near side of the restricted area Ze based on the image data 30a and identifies the near edge E1 of the restricted area Ze based on the restriction material. Furthermore, the control unit 20 acquires a first distance Db1 that defines the first detection range based on the parameter information 30b. The control unit 20 then identifies within the image the range of the first distance Db1 from the near edge E1 of the restricted area Ze, which is delimited by the restriction material, and sets this range as the first detection range Z1.

[0031] The second detection range is a range on the lane L1 that is a second distance from the restriction area Ze to the near side. Therefore, the control unit 20 references the restriction material identified based on the image data 30a and identifies the near end E2 of the restriction area Ze on the lane L1. Furthermore, the control unit 20 acquires a second distance Db2 that defines the second detection range based on the parameter information 30b. The control unit 20 then identifies within the image the range from the near end E2 of the restriction area Ze, which is separated by the restriction material, to the second distance Db2, and sets this range as the second detection range Z2.

[0032] Next, the control unit 20 sets the thresholds (step S105). That is, the control unit 20 refers to the parameter information 30b and acquires the first threshold V1, a value for determining the time threshold (0.5 in this embodiment), and the second threshold V2.

[0033] In this embodiment, camera 40 captures images at a predetermined frame rate. After capturing the number of frames required to calculate the vehicle speed, control unit 20 executes the process of steps S110 to S155 for each frame of image.

[0034] After step S110, first, the control unit 20 acquires images using the function of the image acquisition unit 21a (step S110). That is, the control unit 20 stores the images output by the camera 40 as image data 30a in the recording medium 30. Note that images across multiple frame rates are stored in a state in which the chronological order of the images can be identified, for example, as a video file.

[0035] Next, the control unit 20 detects a vehicle using the function of the vehicle detection unit 21b (step S115). That is, the control unit 20 detects an image of a vehicle included in the image acquired in step S110. This process can be performed by various methods, for example, pattern matching processing. The control unit 20 assigns identification information to the acquired vehicle. However, if a vehicle is also detected and assigned identification information in the frame captured immediately before, and the characteristics of the vehicle detected in both frames are identical, the same identification information is assigned. Various conditions may be used to determine whether the vehicle in the frame captured immediately before is the same as the vehicle in the latest frame. For example, whether the vehicles are the same may be determined on the condition that the distance traveled by the vehicle between frames is within a reasonable range.

[0036] Next, the control unit 20 detects the vehicle position and speed using the function of the warning target vehicle detection unit 21c (step S120). Specifically, the control unit 20 identifies the position in the image of the vehicle detected in step S115. The control unit 20 also compares the position of the vehicle detected in each of the images acquired in step S110 with the image captured immediately before that image, and converts the change in position into a distance on the road. The control unit 20 then identifies the time required to travel that distance from the frame interval, and obtains the vehicle speed by dividing the distance by the time. Various methods can be used to obtain the vehicle speed. For example, the vehicle speed may be determined based on multiple frames of images.

[0037] Next, the control unit 20 determines whether or not the vehicle detected in step S115 is present within the first detection range by using the function of the warning target vehicle detection unit 21c (step S125). That is, if the position in the image of the vehicle detected in step S120 is included in the first detection range set in step S100, the control unit 20 determines that the vehicle is present within the first detection range.

[0038] In this embodiment, when a vehicle is traveling within the first detection range, it is considered to be traveling on boundary line Lb1. Various criteria may be used to determine whether a vehicle is traveling within the first detection range. For example, a configuration may be adopted in which a vehicle is determined to be traveling within the first detection range when at least a portion of the vehicle image overlaps the first detection range. If it is determined that a vehicle is traveling within the first detection range, the control unit 20 associates information (such as a flag) indicating that the vehicle is traveling within the first detection range, i.e., traveling on a boundary line, with the vehicle position acquired in step S120. In this case, the control unit 20 executes steps S130 and subsequent steps. If it is not determined that the vehicle is traveling within the first detection range, the control unit 20 does not consider the vehicle to be traveling within the first detection range. In this case, the control unit 20 executes steps S140 and subsequent steps.

[0039] Next, the control unit 20 determines whether the speed of the vehicle is equal to or greater than the first threshold value V1 and whether the length of time that the vehicle has been traveling within the first detection range is equal to or greater than the time threshold value (step S130) by using the function of the warning target vehicle detection unit 21c. That is, the control unit 20 determines whether the speed acquired in step S120 is equal to or greater than the first threshold value V1.

[0040] Furthermore, the control unit 20 determines a time threshold based on the value for determining the time threshold acquired in step S105. In this embodiment, the time threshold is a value obtained by multiplying the length of time it takes for the vehicle to travel along the boundary line within the first detection range by the value for determining the time threshold (0.5 in this embodiment). By defining the time threshold in this manner, it is possible to define the time threshold as a value for determining whether the vehicle is traveling along the boundary line for a time length equal to or greater than a predetermined percentage of the length of time required for the vehicle to travel the entire length of the first detection range. Furthermore, even if the speed of vehicles traveling through the first detection range varies from vehicle to vehicle, it is possible to define a time threshold that corresponds to changes in speed.

[0041] FIG. 4 is a diagram illustrating the change in the time threshold value depending on the vehicle speed. In FIG. 4, the horizontal axis represents time and the vertical axis represents distance, and the graph shows the relationship between time and distance when the vehicle travels through the first detection range at speeds Va and Vb. It is assumed here that the vehicle enters the first detection range from the rear end in the traveling direction (end E0 shown in FIG. 2) and travels along boundary line Lb1 to the frontmost end (end E1 shown in FIG. 2). That is, the relationship between time and distance when the vehicle enters end E0 and travels to end E1 at a constant speed Va is shown by the solid line. The time length Ta shown in FIG. 4 is the time length required to travel the first distance Db1 at speed Va. The relationship between time and distance when the vehicle travels at speed Vb is shown by the dashed line. The time length Tb shown in FIG. 4 is the time length required to travel the first distance Db1 at speed Vb. The time threshold is calculated by multiplying the time length required for the vehicle to travel the first distance Db1 by 0.5, a value for determining the time threshold. Therefore, for the speed Va, the time threshold is Ta / 2, which is half the time length Ta. For the speed Vb, the time threshold is Tb / 2.

[0042] Based on the above-described concept, the control unit 20 acquires the time threshold. Note that since the time threshold depends on the vehicle speed, if the vehicle speed changes during the loop of steps S110 to S155, the time threshold may be calculated based on the latest speed and changed, or may be fixed to a time threshold acquired based on the speed when the vehicle first entered the first detection range.

[0043] Furthermore, the control unit 20 determines the length of time that the vehicle has been traveling within the first detection range. Specifically, the control unit 20 determines the position of a vehicle associated with the same identification information as the vehicle identification information acquired in step S115, and associated with information indicating that the vehicle is traveling within the first detection range. If the information indicating that the vehicle is traveling within the first detection range is associated with the position of the vehicle, the control unit 20 determines that the vehicle has been traveling within the first detection range at that position. Furthermore, the control unit 20 determines that the vehicle has been traveling within the first detection range for a frame interval (e.g., 1 / 60 seconds if the frame rate is 60 fps) at that position. Then, the control unit 20 determines the length of time that the vehicle has been traveling within the first detection range by accumulating the frame intervals at each position.

[0044] The above process makes it possible to obtain the length of time when a vehicle enters the first detection range and continues to travel within the first detection range. FIG. 5 shows an example in which the boundary line Lb1 shown in FIG. 2 is extracted and the vehicle travels within the first detection range Z1. In FIG. 5, the vehicle's travel trajectory is shown as trajectory O1. Also, in FIG. 5, below the boundary line Lb1, a chart is shown on the time axis indicating whether the vehicle traveled within the first detection range or outside the first detection range.

[0045] 5, it is assumed that a vehicle enters the first detection range Z1 at time T1 and travels within the first detection range Z1 until time T2. In this case, the control unit 20 acquires a time length T21 corresponding to the period T2-T1, and determines that the vehicle has traveled within the first detection range Z1 for the time length T21.

[0046] Fig. 6 illustrates a case in which the vehicle travels along a different trajectory O2 from Fig. 5 within the same first detection range Z1 as Fig. 5. In the example illustrated in Fig. 6, the vehicle is meandering, and the vehicle enters first detection range Z1 during the period from time T3 to time T4 and the period from time T5 to time T6, and travels outside first detection range Z1 during the other periods. In this case, control unit 20 obtains a time length T43 corresponding to the period from T4 to T3 and a time length T65 corresponding to the period from T6 to T5, and obtains the sum of the obtained time lengths T43 and T65 as the length of time that the vehicle traveled within first detection range Z1.

[0047] When the length of time that the vehicle has been traveling within the first detection range Z1 is acquired in the above manner, the control unit 20 determines whether the length of time is equal to or greater than the above-mentioned time threshold. If it is determined in step S130 through the above processing that the vehicle speed is equal to or greater than the first threshold V1 and the length of time that the vehicle has been traveling within the first detection range is equal to or greater than the time threshold, the control unit 20 acquires the vehicle as a first vehicle to be warned by using the function of the warning target vehicle detection unit 21c (step S135).

[0048] On the other hand, if it is determined in step S130 that the vehicle speed is equal to or greater than the first threshold V1 and that the length of time the vehicle has been traveling within the first detection range is not equal to or greater than the time threshold, the control unit 20 executes the processes from step S140 onward. That is, if the speed of the vehicle traveling within the first detection range is not equal to or greater than the first threshold V1, the vehicle is decelerating before entering the restricted area Ze, and therefore is not subject to a warning in this embodiment. Also, if the length of time the vehicle has been traveling within the first detection range is not equal to or greater than the time threshold, the vehicle is considered to have temporarily entered the first detection range Z1 but has since returned to lane L1 and is not subject to a warning in this embodiment.

[0049] In this way, if the vehicle detected in step S115 is not the first vehicle to be warned, the control unit 20 determines whether the vehicle detected in step S115 is present within the second detection range using the function of the warning vehicle detection unit 21c (step S140). That is, if the position in the image of the vehicle detected in step S120 is included in the second detection range set in step S100, the control unit 20 determines that the vehicle is present within the second detection range. If it is not determined in step S140 that the vehicle is present within the second detection range, the control unit 20 repeats the processes from step S110 onwards for the next frame captured by the camera 40.

[0050] On the other hand, if it is determined in step S140 that the vehicle is present within the second detection range, the control unit 20 determines whether the vehicle speed is equal to or greater than the second threshold value V2 using the function of the warning target vehicle detection unit 21c (step S145). That is, the control unit 20 determines whether the speed acquired in step S120 is equal to or greater than the second threshold value V2. If it is determined in step S145 that the vehicle speed is equal to or greater than the second threshold value V2, the control unit 20 acquires the vehicle as a second warning target vehicle using the function of the warning target vehicle detection unit 21c (step S150). If it is not determined in step S145 that the vehicle speed is equal to or greater than the second threshold value V2, the control unit 20 repeats the processes from step S110 onwards for the next frame captured by the camera 40.

[0051] When step S135 or step S150 is executed, the control unit 20 causes the warning unit 50 to output warning information using the function of the warning control unit 21d (step S155). That is, the control unit 20 controls the warning unit 50 to output a warning sound. When step S155 is executed, the control unit 20 repeats the processes from step S110 onwards for the next frame captured by the camera 40. With the above configuration, it is possible to warn of vehicles that may approach the restricted area Ze.

[0052] Furthermore, according to this embodiment, a warning is issued when a first warning target vehicle is detected, so that when a vehicle is heading toward the restricted area Ze, not exhibiting normal behavior such as staying within the lane but exhibiting abnormal behavior such as staying on the boundary line for a long time, a warning can be given to an operator in the restricted area Ze. Such abnormal behavior occurs when the driver of the vehicle is dozing off or distracted while driving, and this embodiment makes it possible to warn an operator of the presence of a vehicle exhibiting such abnormal behavior.

[0053] Furthermore, in this embodiment, a first warning target vehicle is detected from among vehicles traveling within the first detection range, and therefore, whether a vehicle is a first warning target vehicle can be determined by simpler processing than in a configuration in which vehicles exhibiting abnormal behavior are detected from the entire image.

[0054] Furthermore, in this embodiment, the time threshold value is a value obtained by multiplying the length of time during which the vehicle travels along the boundary line within the first detection range by a value greater than 0 and less than 1. According to this configuration, it is possible to determine whether or not the vehicle's travel along the boundary line is abnormal using a simple determination criterion.

[0055] Furthermore, according to this embodiment, a warning is issued when a second vehicle requiring warning is detected, so that a warning can be issued for a vehicle approaching the restricted area Ze at an excessive speed. Therefore, the operator can be warned of the presence of a vehicle whose driver may not be aware of the restricted area Ze.

[0056] (3) Other embodiments: The above embodiment is one example for implementing the present invention, and various other embodiments are also possible. For example, the vehicle warning system 10 may be realized by a plurality of devices. The vehicle warning system 10 may also include a detection device that detects a vehicle to be warned and a warning device that outputs a warning. Some of the components of the above embodiment may be omitted, the order of processing may be changed or omitted, or processing may be replaced with an alternative process.

[0057] In the above-described embodiment, step S100 of setting the first detection range and the second detection range may be performed in various ways. For example, based on an image captured by camera 40, a user may use an input unit to specify an area within the image to indicate the first detection range and the second detection range.

[0058] Furthermore, in the above-described embodiment, when a first vehicle to be warned is present, a warning is output without considering the presence of a second vehicle to be warned. However, when a first vehicle to be warned is present, it may also be determined whether a second vehicle to be warned is present. In this case, a warning is output if at least one of the first vehicle to be warned and the second vehicle to be warned is present. Furthermore, when both a first vehicle to be warned and a second vehicle to be warned are present, the strength of the warning may be increased compared to when either one of them is present.

[0059] Furthermore, in the above-described embodiment, whether or not a vehicle is a vehicle subject to the first warning is determined based on the behavior of the vehicle on the solid boundary line Lb1 at the edge of the road in the width direction, which is located on the left side when the vehicle faces forward in the direction of travel just before the restricted area Ze, but this configuration is not limited to this. For example, whether or not a vehicle is a vehicle subject to the first warning may be determined based on the behavior of the vehicle on the lane boundary line Lb2 located on the right side when the vehicle faces forward in the direction of travel just before the restricted area Ze.

[0060] Furthermore, the conditions for identifying the first and second warning target vehicles are not limited to those in the above-described embodiment. For example, in step S130, a vehicle may be deemed to be the first warning target vehicle if either the speed of the vehicle within the first detection range is equal to or greater than the first threshold value V1 or the length of time the vehicle has been traveling within the first detection range is equal to or greater than the time threshold value.

[0061] Furthermore, the time threshold is not limited to the above-described embodiment. That is, the time threshold may be a value that can be used to determine whether a vehicle traveling within the first detection range Z1 for a period of time equal to or longer than the time threshold is abnormal. Therefore, the time threshold may be determined using various methods. For example, the time threshold may be determined based on the vehicle speed. Furthermore, in this case, the time threshold may be configured to be shorter as the vehicle speed increases. With this configuration, even when the vehicle speed changes, a time threshold that can determine whether the behavior is abnormal in response to changes in speed can be defined. Furthermore, the threshold may be defined by distance instead of time. For example, a certain percentage (0 to 100%) of the total length of the first detection range Z1 may be used as the threshold, and the distance traveled by the vehicle within the first detection range Z1 may be compared with the threshold. In this case, the threshold may also be configured to be smaller as the vehicle speed increases.

[0062] The image acquisition unit only needs to be able to acquire an image of the area in front of the restricted area set on the road. In other words, the image acquisition unit only needs to be able to acquire an image of the area in which a vehicle that may be subject to a warning is traveling. A restricted area is an area on the road where traveling is prohibited. Therefore, various restricted areas may exist other than construction sites. For example, a restricted area may be an accident site, an area where traveling of vehicles is prohibited due to various investigations, events, etc.

[0063] The area in front of the restricted area is an area that exists behind the restricted area in the direction of travel of the road. The area in front of the restricted area may be an area in which a vehicle that may be the target of a warning travels, and may be an area that has a predetermined positional relationship with the restricted area. The image may be acquired so that it can include the vehicle, and various configurations may be adopted other than the configuration in which the image is captured by a camera installed on a work vehicle as described above. For example, the image may be acquired by attaching various sensors such as cameras to restricting materials placed on the road, posts installed on the road, or beams extending from the posts. Furthermore, the image may be a still image or a video image as long as it is used to detect vehicles heading toward the restricted area.

[0064] The vehicle detection unit is only required to be able to detect a vehicle traveling toward a restricted area based on an image. In other words, if a vehicle traveling toward a restricted area is present in an area before the restricted area, an image of the vehicle will be included in the image. Therefore, the vehicle detection unit is only required to be able to detect the vehicle by analyzing the image. The analysis is not limited to the above-described method. For example, a configuration may be adopted in which a machine learning model that identifies whether each object in an image is a vehicle is created in advance, and the vehicle is detected by inputting the image into the machine learning model. The machine learning model can be generated using various known methods.

[0065] The warning target vehicle detection unit is only required to be able to detect a vehicle traveling on a lane boundary line set on a road at a speed equal to or greater than a first threshold as a first warning target vehicle. Furthermore, the warning target vehicle detection unit is only required to be able to detect a vehicle traveling at a speed equal to or greater than the second threshold within a second detection range that is a predetermined distance in front of the restricted area as a second warning target vehicle. In other words, the warning target vehicle detection unit detects a warning target vehicle based on whether the behavior of the vehicle detected by the vehicle detection unit is a predetermined warning target behavior. The warning target behavior may be various behaviors, and a warning target vehicle may be a vehicle whose behavior is different from the behavior of the first warning target vehicle and the second warning target vehicle.

[0066] A first warning target vehicle is a vehicle whose behavior can be considered abnormal (or for which a warning should be issued) as a vehicle behavior on a normal road. In other words, a vehicle is detected as a first warning target vehicle when it deviates from normal behavior on a normal road. Various factors may be used to determine whether a vehicle is a first warning target vehicle. In the above-described embodiment, the conditions for determining whether a vehicle is a first warning target vehicle are that the vehicle is traveling on a boundary line, that the speed is equal to or greater than a first threshold, that the length of time the vehicle travels on the boundary line is equal to or greater than a time threshold, and that the vehicle travels within a first detection range, but these are not limited to these. In other words, a vehicle that satisfies at least one of the conditions of the traveling position, the speed, the length of time the vehicle is in a specific area, and that the vehicle is traveling within a specific area may be a first warning target vehicle, or a vehicle that satisfies other conditions may be a first warning target vehicle. Examples of other conditions include sudden acceleration / deceleration, sudden steering, and other behaviors of the vehicle.

[0067] Lane boundaries are dashed or solid lines that separate lanes, and since vehicles are usually expected to travel between the boundaries, if there is a vehicle traveling on a boundary line, that vehicle can be set as a warning target. Of course, behaviors such as traveling on a boundary line for a short period of time to change lanes or move to the shoulder of the road may be excluded from the warning target.

[0068] A vehicle subject to the second warning is a vehicle whose behavior can be considered abnormal (or should be warned about) based on its positional relationship with the restricted area. In other words, if the vehicle's speed does not decrease sufficiently before the restricted area, it can be assumed that some abnormality is occurring, such as the driver not noticing the restricted area or falling asleep. Therefore, if the vehicle is traveling at an excessive speed within a second detection range of a predetermined distance before the restricted area in a lane where the restricted area exists, which is equal to a second threshold, it can be assumed that the vehicle is behaving abnormally. Various factors may be used to determine whether or not a vehicle is subject to the second warning. For example, a vehicle may be deemed to be subject to the second warning if the distance from the restricted area is equal to or less than a threshold.

[0069] The warning control unit is only required to be able to cause the warning unit to output warning information when a first vehicle to be warned is detected and when a second vehicle to be warned is detected. In other words, the warning control unit is only required to be able to warn of the presence of a vehicle that may enter the restricted area. The warning mode is not limited to the configuration of outputting a warning sound to a worker working in the restricted area, as described above. For example, the warning may be given by displaying characters or the like indicating a warning on a sign present in or around the restricted area. Alternatively, the warning may be given by transmitting a signal wirelessly or otherwise to a device carried by the worker, causing the device to output a sound, an image, a vibration, or the like.

[0070] The present invention can also be applied as a program or method. It can also be modified as needed, such as being partly software and partly hardware. Furthermore, the invention can also be implemented as a recording medium for a program that controls a device. Of course, the recording medium for the software can be a magnetic recording medium or a semiconductor memory, and any recording medium developed in the future can be considered in exactly the same way. [Explanation of symbols]

[0071] 10...vehicle warning system, 20...control unit, 21...vehicle warning program, 21a...image acquisition unit, 21b...vehicle detection unit, 21c...warning target vehicle detection unit, 21d...warning control unit, 30...recording medium, 30a...image data, 30b...parameter information, 40...camera, 50...warning unit

Claims

1. an image acquisition unit that acquires an image of an area in front of a restricted area set on a road; a vehicle detection unit that detects a vehicle traveling toward the restricted area based on the image; a warning target vehicle detection unit that detects, as a first warning target vehicle, a vehicle that travels on a lane boundary line set on the road at a speed equal to or greater than a first threshold value and that travels on the boundary line for a length of time equal to or greater than a time threshold value; a warning control unit that causes a warning unit to output warning information when the first warning target vehicle is detected; A vehicle warning system comprising:

2. The warning target vehicle detection unit detecting the first vehicle to be warned from among the vehicles traveling within a first detection range of a predetermined distance along the boundary line from the restricted area toward the front side; The vehicle warning system of claim 1 .

3. the time threshold is a value obtained by multiplying a time length during which the vehicle travels along the boundary line in the first detection range by a value greater than 0 and less than 1; 3. The vehicle warning system of claim 2.

4. The time threshold is shorter as the speed of the vehicle is higher. The vehicle warning system of claim 1 .

5. The warning target vehicle detection unit a second detection range that is a predetermined distance in front of the restricted area on the lane, and the vehicle traveling at a speed equal to or greater than a second threshold is detected as a vehicle to be warned; The warning control unit When the second vehicle to be warned is detected, the warning unit outputs the warning information. The vehicle warning system according to any one of claims 1 to 4.

6. The inside and outside of the restricted area are separated by restrictive materials placed on the road, The image acquisition unit acquires the image taken by a camera installed on the restriction material or a work vehicle located within the restriction area. The vehicle warning system according to any one of claims 1 to 4.

7. A step in which an image acquisition unit acquires an image of an area in front of a restricted area set on a road; a step in which a vehicle detection unit detects a vehicle traveling toward the restricted area based on the image; a step in which a warning target vehicle detection unit detects, as a first warning target vehicle, a vehicle that travels on a lane boundary line set on the road at a speed equal to or greater than a first threshold value and that travels on the boundary line for a length of time equal to or greater than a time threshold value; a warning control unit causing a warning unit to output warning information when the first warning target vehicle is detected; A vehicle warning method implemented in a vehicle warning system, comprising:

8. Computer, an image acquisition unit that acquires an image of an area in front of a restricted area set on a road; a vehicle detection unit that detects a vehicle traveling toward the restricted area based on the image; a warning target vehicle detection unit that detects, as a first warning target vehicle, the vehicle that travels on a lane boundary line set on the road at a speed equal to or greater than a first threshold value and that travels on the boundary line for a length of time equal to or greater than a time threshold value; a warning control unit that causes a warning unit to output warning information when the first warning target vehicle is detected; This vehicle warning program acts as a

Citation Information

Patent Citations

  • Traffic monitoring and warning system

    JP2006291502A

  • Moving Object Detection System

    JP6980887B1

  • Advanced warning and risk evasion system and method

    US20150091716A1