JP3218834B2 - Obstacle detection device for vehicles - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a method for automatically controlling a vehicle, performing a safe driving operation, or performing a preceding vehicle following automatic speed control. The present invention relates to a vehicle obstacle detection device that determines a traveling lane based on the traveling lane.

[0002]

2. Description of the Related Art As a conventional obstacle detecting device for a vehicle, for example, a "traveling road detecting device" disclosed in Japanese Patent Application Laid-Open No. 4-36878 or a Japanese Patent Application Laid-Open No. Sho 62-273377.
There is known a "vehicle surrounding monitoring device" disclosed in Japanese Unexamined Patent Application Publication No. H11-163,873.

[0003] A "traveling path detecting device" disclosed in Japanese Patent Application Laid-Open No. 4-36878 detects an edge point inside the traveling path after detecting the traveling path, and the edge lump is continuously detected for a certain period of time. When it is determined that the object exists, it is detected as an obstacle.

A "vehicle surrounding monitoring apparatus" disclosed in Japanese Patent Laid-Open Publication No. Sho 62-273377 rotates an ultrasonic sensor and an infrared imaging camera according to the running state of a vehicle, and is located in front of the vehicle. It determines the existence of a living body such as a pedestrian and notifies the driver.

Further, in addition to the above, like a laser radar, light is emitted and its reflected light is received, and the distance to a preceding vehicle or an obstacle is detected based on the round-trip time difference between light emission and light reception. Many devices are also known.

[0006]

However, the conventional apparatus as described above has the following problems. That is, in the "traveling path detecting device" disclosed in Japanese Patent Application Laid-Open No. 4-36878, when image processing is performed, detection simply depends on the presence or absence of an edge. However, characters and patterns drawn on the road surface are erroneously detected as obstacles, and there are many errors in distance measurement.

Further, the "vehicle surroundings monitoring device" disclosed in Japanese Patent Application Laid-Open No. 62-273377 has a problem that high-speed processing cannot be performed because the entire captured image is processed as a detection target. . Further, when using the above laser radar, in a curve on a traveling road, a vehicle in another lane or a reflector installed on a road side zone, etc.
There is a problem in that it is detected as a forward obstacle.

The present invention has been made in view of the above, and specifies a road surface area necessary for an alarm or control, executes processing such as the presence or absence of an obstacle only in the area, and performs an obstacle detection process. It is an object of the present invention to reduce and simplify the amount of processing and realize high-speed processing and low cost of the apparatus.

[0009]

SUMMARY OF THE INVENTION In order to achieve the above object, the present invention provides a distance measuring means for measuring a distance to an obstacle in front of a host vehicle, and image information obtained by imaging a state in front of the host vehicle. Inputting a captured image , and measuring the distance measuring means from the captured image from the captured image inputting means
A road surface area extracting unit that extracts a road surface region that is a range, a road surface detecting unit that detects a straight line formula of a boundary of the road surface region extracted by the road surface region extracting unit, and a straight line formula of a boundary that is detected by the road surface detecting unit First determining means for determining a traveling lane based on the distance to the obstacle measured by the distance measuring means, and second determining whether or not the obstacle is present in the traveling lane. Judgment means.

[0010]

SUMMARY OF vehicular obstacle detection system according to the present invention, the distance
Measure the distance to the obstacle in front of the vehicle by measuring means.
And captures the state in front of the vehicle using the captured image input means.
The captured image is input, and the distance measurement unit measures the captured image.
The road surface area which is the range is extracted by the road surface area extracting means,
Furthermore, the straight line at the boundary of the road surface area is represented by the road surface extraction means.
Detects a straight line expression that is a mathematical expression, and measures the straight line expression and distance of the boundary.
Based on the distance measured by the means, the obstacle
Because it is determined whether or not it is on the line,
Image processing area for issuing an obstacle warning ahead of the lane
It is limited to only the measurement range of the distance measuring means.

[0011]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of an obstacle detecting device for a vehicle according to the present invention. FIG. 1 is a block diagram showing a schematic configuration of an obstacle detecting device for a vehicle according to the present embodiment. The optical radar 101 is installed in front of the vehicle and is installed in front of the vehicle similarly to an optical radar 101 using a laser radar. An image input unit 102 for inputting a captured image in front of the vehicle, processing an image signal from the image input unit 102, and integrating the signal with the signal from the optical radar 101;
The image processing unit 103 includes an image processing unit 103 that determines whether there is a preceding vehicle, and an alarm / control unit 104 that performs automatic control such as alarm output and brake control.

Further, the image processing unit 103 includes a window processing unit 1 for inputting image data from the image input unit 102.
05 and a road surface detecting unit 106, and a judging unit 108 that outputs control information to the alarm / control unit 104 based on the output from the optical radar 101 and the output from the road surface detecting unit 106. It is composed of

Next, the operation will be described. Optical radar 1
The information on the distance between the host vehicle and the obstacle ahead of the host vehicle detected by 01 is input to the determination unit 108 in the image processing unit 103. At the same time, the image information obtained by imaging the front of the vehicle is
The image is input from the image input unit 102 to the window processing unit 105 in the image processing unit 103. In the window processing unit 105, a road surface area (window)
And outputs this to the road surface detection unit 106. After that, the road surface detection unit 106 executes a later-described boundary linear expression detection process, and outputs the result to the determination unit 108. The determination unit 108 determines the traveling lane based on the distance information to the obstacle and the straight line formula of the boundary, and outputs the determination result to the alarm / control unit 104 as a control signal. The alarm / control unit 104 executes an alarm output, deceleration control, and the like based on the determination result from the determination unit 104.

Next, the operations of the window processing unit 105, the road surface detection unit 106, the judgment unit 108, and the alarm / control unit 104 in this embodiment will be described in detail. First, the window processing unit 105 cuts out only the measurement range of the optical radar 101 from an image input from the image input unit 102 as a window (a detection window of a road surface area). FIG.
FIG. 7 is an explanatory diagram showing a state of window cutout by the window processing unit 103, in which a hatched portion is a window 2
01. In this case, it is assumed that the optical radar 101 and the captured image input unit (image input unit 102) are both fixed at predetermined positions in front of the vehicle. And the optical radar 1
Assuming that the measurement range (measurement area) of 01 is known, the measurement range of the optical radar 101 in the captured image, that is, the window 201 can be fixedly determined. Therefore, the following processing is performed in the above window 20
It is performed within the range of 1.

In the road surface detecting section 106, the window 201
, Other than the road surface such as a white line, a preceding vehicle, and an obstacle are detected. FIG. 3 is an explanatory diagram showing an example of detection of a road surface area by the road surface detection unit 106. First, an edge is detected in the window 201, and then a straight line is detected. For example, in FIG. 3A, since the white line is longer than the straight line of the contour of the preceding vehicle, the white line is output as a detection straight line. On the other hand, in FIG. 3B, since the white line is hidden by the preceding vehicle, the horizontal edge of the preceding vehicle is output as a detection straight line.

The detection straight line is such that the x-axis is a horizontal axis (a direction orthogonal to the traveling direction) and the y-axis is a vertical axis (the same direction as the traveling direction).
, A and b are output according to x = a · y + b. The y-axis passes through the center of the window 201.

Next, the operation of the judgment unit 108 will be described. Assuming that the distance measured by the optical radar 101 is D, a point on the road surface in front D can be represented on the image by the equation _yD = F.H / D. Note that F is the focal length of the lens, H is the height of the camera, and all values are known.

[0018] When compared with the results of the detection straight line from the road surface detecting unit 106, if the detected straight lines if the lower portion of the preceding vehicle becomes substantially equal to the y _D and b. Also,
Since the horizontal edge of the preceding vehicle is likely to be detected,
As shown in FIG. 3B, a becomes substantially zero. On the other hand, as shown in FIG. 3A, if the preceding vehicle is traveling on the left lane of the own lane in the left curve, a is 0.
A large value as compared to, b is a considerably smaller value for y _D.

From the above, the measurement result of the optical radar 101 is compared with the above-mentioned detection straight line result. As a result, when a is almost 0, the preceding vehicle detected by the optical radar 101 is in the own lane. It is determined that it is located ahead, and the result is output to the alarm / control unit 104. Optical radar 1
If no obstacle is detected in 01, it is determined that there is no preceding vehicle. On the other hand, the road surface detection unit 106 does not detect the straight line, and
If the preceding vehicle is detected by the method 1, it is determined that there is a vehicle having the same density (same color) as the road surface or that the road surface has not been detected due to backlight or the like. to decide.

The alarm / control unit 104 controls to issue an alarm or decelerate, for example, in a situation where a preceding vehicle is present in front of the own lane and is approaching. Note that it is also possible to determine whether the road is a right curve or a left curve from the value of a. further,
The magnitude of the road curve can be known from the magnitude of the absolute value of a (a sharp curve when a is small). For example,
When the vehicle speed is high and the curve is tight, it is possible to issue an alarm or control the vehicle to decelerate.

Next, another embodiment of the road detector 106 will be described. In this embodiment, a color camera is used as a captured image input means because color is used instead of edge detection or straight line detection.

FIG. 4 is a block diagram showing the flow of the detection process of the road surface detection unit 106. In the figure, first, the color of the lower end of the window 201 is extracted (S401).
Since the lower end of the window 201 is very short, it is unlikely that the preceding vehicle is imaged. Therefore, the color of the road surface can be detected from the lower end portion of the window 201.

Next, the degree of similarity with the color extracted in the window 201 is calculated, and a similar color (for example, the same color) is calculated.
Are detected (S402). Thereafter, a labeling process (region division) is executed (S403), an area is obtained for each region, and only the region having the maximum area is extracted (S40).
4). Thereafter, the region having the maximum area is projected in parallel with the x-axis, and the frequency for y is obtained (S40).
5). Finally, a peak value is detected from the frequency distribution (S4).
06), the y coordinate giving the peak is output as b.
Alternatively, instead of the peak value, the uppermost y coordinate exceeding a certain threshold is output as b (S40).
7), a straight line can be detected (S408). FIG.
FIG. 6 and FIG. 6 show examples of the respective detections.

That is, in FIG. 5 and FIG.
The same colors are extracted from the image information shown in FIGS. 6A and 6A to obtain the images shown in FIGS. 5B and 6B. Thereafter, labeling and extraction of the maximum area are performed on the images shown in FIGS. 5B and 6B, and the images shown in FIGS.
The image shown in (c) is obtained. 5 (c) and FIG.
The images shown in FIGS. 5D and 6D are obtained by extracting the maximum y coordinate and detecting the straight line from the image shown in FIG. On the other hand, the images shown in FIGS. 5 (e) and 6 (e) are obtained from the images shown in FIGS. 5 (c) and 6 (c) by y-axis projection and peak detection.

Further, the determination unit 108, when the b and y _D is approximately equal, or when it is b> y _D, it is determined that there is the preceding vehicle in the own traffic lane, while next time is b <y _D It is determined that the preceding vehicle exists in the lane.

In this embodiment, a portion having the same color as that of the road surface is detected, and the road surface region is interrupted by a white line or a preceding vehicle. Further, since the road surface region having the largest area is considered to be a region including immediately before the vehicle, a method of examining how far the region is continuous far away is used. By using this method, it is possible to detect a road surface area with less influence of a step or a shadow on the road surface, so that there is an advantage that the accuracy of information is improved.

As described above, as shown in FIGS. 5 and 6, the road surface area is extracted from the imaged image, the straight line expression between the road surface region and the boundary is detected, and the straight line expression of the boundary and the optical radar 101 are detected. The traveling lane is determined based on the distance detected by (1). Therefore, the area required for the alarm,
Processing only needs to be performed for the measurement area of the optical radar 101, and high-speed processing is realized because the processing amount is small.

From the region having the maximum area in the present embodiment, the maximum y coordinate with respect to the x axis is obtained.
By performing the straight line detection, similarly to the above-described embodiment, the degree of the curve or the right or left curve can be well known.

As described above, the distance measurement result by the optical radar 101 and the processing result of the captured image by the image input unit 102 are integrated, and only the area corresponding to the measurement range (window 201) of the optical radar 101 in the image is obtained. By executing the extraction processing of non-road areas or road areas such as white lines and obstacles, it becomes possible to detect the preceding vehicle ahead of the own lane. In addition, a simple configuration and high-speed processing can be performed. As a result, the cost of the apparatus can be reduced.

[0030]

As described above, according to the obstacle detecting device for a vehicle according to the present invention, a captured screen is input from the captured image input means, and a road surface area is extracted from the screen.
Further, the road surface area extracting means detects the straight line formula of the boundary of the road surface area, and determines the traveling lane based on the straight line formula of the boundary and the distance measured by the distance measuring means. Since the processing is executed only for the measurement area of the distance measuring means, the processing amount of the obstacle detection processing can be reduced and simplified, and high-speed processing and lower cost of the apparatus can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of a vehicle obstacle detection device according to the present invention.

FIG. 2 is an explanatory diagram showing a state of window cutout by a window processing unit shown in FIG. 1;

FIG. 3 is an explanatory diagram illustrating an example of detection of a road surface area by a road surface detection unit illustrated in FIG. 1;

FIG. 4 is a block diagram illustrating a flow of a road surface area detection process performed by a road surface detection unit illustrated in FIG. 1;

FIG. 5 is an explanatory diagram showing an example of a straight line type detection process of a road surface area and a boundary according to the present invention.

FIG. 6 is an explanatory diagram showing an example of a straight-line detection process of a road surface area and a boundary according to the present invention.

[Explanation of symbols]

 Reference Signs List 101 optical radar 102 image input unit 103 image processing unit 104 alarm / control unit 105 window processing unit 106 road surface detection unit 107 road surface detection unit in window 108 judgment unit

Claims (1)

(57) [Claims] 1. A distance measuring means for measuring a distance to an obstacle in front of a host vehicle, a picked-up image input means for inputting image information obtained by picking up an image of a state in front of the host vehicle, and an image pick-up from the picked-up image input means Measurement range of the distance measurement means from the image
And the road region extracting means for extracting a road area is, and the road surface detecting means for detecting a linear equation of the boundary of the road area extracted by the road region extracting means, a linear equation of the detected boundary by the road surface detecting means First determining means for determining a traveling lane based on the distance to the obstacle measured by the distance measuring means, and second determining for determining whether the obstacle exists in the traveling lane Means for detecting a vehicle obstacle.