JPH04313014A - Car interval inspecting device - Google Patents

Abstract

PURPOSE:To stably measure the distance to a target after the target is once fixed and, at the same time, to confirm the preceding car which is used as the car interval measuring target even when a plurality of preceding cars exist. CONSTITUTION:This car interval detecting device is constituted of two image sensors 3 and 4, displaying means 11 which displays the image of a preceding car 5 taken by means of one of the sensors, picture tracking means 12 which tracks the image of the car 5 by using a window surrounding the image of the car 5, binarization processing means 14 which extracts the edge component of an input picture, window position correcting means 15 which discriminates the symmetry of a binarized picture obtained by means of the means 14 and corrects the position of the window surrounding the image of the car 5, and distance detecting means which calculates the distance to the object surrounded by the window. Therefore, the distance to the preceding car 5 being tracked can be accurately measured and, in addition, even when a plurality of preceding cars exist, the target car 5 can be easily recognized and, in addition, the image of the car 5 can be stably tracked.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical distance detection device using an image sensor, and more particularly to a distance detection device that continuously measures the distance between a preceding vehicle and the own vehicle.

0002

2. Description of the Related Art FIG. 8 is a block diagram of an optical distance detecting device using a conventional image sensor disclosed in, for example, Japanese Patent Publication No. 63-38085 and Japanese Patent Publication No. 63-46363. , 1 and 2 are left and right lenses as an optical system arranged apart by a base line length L, 3 and 4 are image sensors arranged at the focal length f of the lenses 1 and 2, respectively, and 30 is an image sensor 3. ，
This is a signal processing device that calculates the distance to the object 31 using the image signal sent from 4.

Next, the principle of detecting the distance to an object will be explained. Image sensors 3 and 4 are connected through lenses 1 and 2, respectively.
The image signal of the object 31 imaged on the signal processing device 30
The distance R to the object 31 is determined by the following equation based on the principle of triangulation from the shift amount a when the above two image signals match best.

0004

[Math 1]

On the other hand, a method of tracking an image of a preceding vehicle captured by an image sensor, etc.
It is disclosed in Publication No. 52. According to this, the target is image-tracked by an operator setting a tracking gate (window) surrounding the target to be tracked on the display screen while looking at the display screen.

[0006]

[Problems to be Solved by the Invention] Conventional distance detection devices determine the distance to an object by comparing the entire image taken by a pair of left and right or top and bottom optical systems. As a result, the S/N ratio deteriorates, making accurate distance measurement impossible. Additionally, if you try to measure the distance between you and the vehicle in front while you are driving with this device installed on your vehicle, if another vehicle is running on the left or right side of the vehicle in front, you will notice that There was a problem in that the driver did not know whether the system was measuring or not.

[0007] This invention was made to solve the above-mentioned problems, and has a configuration that combines a conventional optical inter-vehicle distance detection device and an image tracking device for the preceding vehicle. By determining the distance to the target object, you can continuously measure the distance to the target object even if it is moving as long as it is within the field of view, and even if there are multiple preceding vehicles, the distance to the target object is measured. An object of the present invention is to provide an inter-vehicle distance detection device capable of confirming which vehicle is a preceding vehicle and stably tracking an image of a preceding vehicle.

[0008]

[Means for Solving the Problems] The inter-vehicle distance detection device according to the present invention compares image signals formed on an image sensor by substantially a pair of upper and lower optical systems, and electrically detects a shift between the two images. An inter-vehicle distance detection device that measures the distance to a preceding vehicle in front of one's own vehicle based on the principle of triangulation includes a display means for displaying an image of a preceding vehicle captured by one image sensor, and a display means surrounding the preceding vehicle image. window and
An image tracking means for image tracking an image of a preceding vehicle using a window, a binarization processing means for extracting edge components of an input image, and a binary image from which edge components have been extracted to determine symmetry within the window. window position correction means for correcting the position of a window surrounding the preceding vehicle image so that a well-symmetrical image is located in the center; and detecting a shift between upper and lower corresponding image signals using an image signal within the window as a reference signal. The present invention is characterized by comprising distance detection means for calculating the inter-vehicle distance between the host vehicle and the preceding vehicle.

[0009]

[Operation] The inter-vehicle distance detection device according to the present invention uses the image signal in the window as a reference signal to detect the deviation between the upper and lower corresponding image signals and calculates the inter-vehicle distance between the own vehicle and the preceding vehicle. The influence of this is reduced and accurate distance measurement is possible. Further, the window position correction means determines symmetry using the binary image from which edge components have been extracted by the binarization processing means, and the image of the preceding vehicle is placed in the center of the window set by the image tracking means. It acts to compensate for the appropriate position.

0010

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows a configuration diagram of an inter-vehicle distance detection device according to an embodiment of the present invention. In FIG. The installed two-dimensional image sensor 5 indicates the preceding vehicle to be followed. 6 is an analog-to-digital converter,
8, 9, 16 are memories, 10 is a microcomputer,
A display screen 11 displays an image captured by the lower image sensor 3, and is controlled by the microcomputer 10. 12 is an image tracking device that forms a window 17 for image tracking of the preceding vehicle image 5a on the display screen 11 as shown in FIG. 3, and also for image tracking of the preceding vehicle image 5a surrounded by the window 17. 1
3 is an image tracking instruction switch operated by the driver. 14 is a binarization processing device that extracts edge components of the image captured by the image sensor 3, and 15 is set by the image tracking device 12 using the binary image obtained by this binarization processing device 14. This is a window position correction device that corrects the set position of the window 17.

Next, the operation will be explained. For example, an image 5a of a preceding vehicle that should follow the image of the lower image sensor 3
Assuming that the window 17 is displayed on the display screen 11 as shown in FIG. The preceding vehicle image 5a moves to a position within the window 17. Then, as shown in Figure 3, the window 17
If the driver operates the image tracking instruction switch 13 when the preceding vehicle image 5a enters the image, the display screen 1
No matter how the leading vehicle image 5a moves within the window 1, the window 17 moves accordingly, and the window 17 automatically tracks the leading vehicle image 5a. In addition,
The image tracking function described above is similar to that of the conventional apparatus disclosed in, for example, Japanese Patent Publication No. 60-33352 or Japanese Patent Publication No. 1-35305, and detailed explanation thereof will be omitted.

Thus, since the image tracking device 12 performs image tracking by comparing images in time series, the window 17 may deviate from the preceding vehicle image 5a due to accumulation of errors, making stable image tracking impossible. By the way, if the object to be measured is a car, the image seen from the rear is
In general, it contains many horizontal edge components and is often symmetrical. Therefore, by utilizing such characteristics, the symmetry is determined and the window setting position is corrected. First, as shown in FIG. 4, in the input image captured by the image sensor 3 and stored in the memory 8, the image tracking device 1
Area 18 surrounded by window 17 set by 2
The processing here involves reading out the image signal of a neighboring region 19 with a width of w pixels to the left and right, and extracting an edge component by the binarization processing device 14. For example, the edge is emphasized using a differential filter, etc. This is like binarizing a differential image at a certain threshold level. The differential binary image obtained through this processing is stored in the memory 16.

Next, the window position correction device 15 judges the symmetry of the binary image stored in the memory 16. It is assumed that the memory 16 stores a binary image as shown in FIG. While shifting a window of a size corresponding to the window 17 on this memory 16, the symmetry is evaluated by performing the calculation shown in the following equation on the image within the window. However, S(i,j) is an image signal of a binary image, and is either 0 or 1. Furthermore, m and n are the width and height of the window 17, respectively, and w is the width of the nearby area 19 stored in the memory 16. Further, it is assumed that k=0, 1, . . . , 2w.

[0014]

[Math 2]

Since this arithmetic expression is for a binary image, it is folded back to the left and right at the central axis of the area within the window, and the logical product of the overlapping portions is calculated. Then, it is determined that the symmetry is best at the position k where the value of Ak is maximum in the above equation 2, and the value of k at that time is determined. Here, if the position of the upper left pixel of the window 17 set by the image tracking device 12 is (p, q) as shown in FIG. 6, then the new position of the upper left pixel is (p-w+k, q). ,
A window 17a having a size of m×n is set, and the image of the area surrounded by the window 17a is used as a reference image in distance calculation. The window setting position is corrected as described above.

Next, a method for detecting the inter-vehicle distance between the host vehicle and the preceding vehicle will be explained. First, microcomputer 10
reads the pixel signal within the corrected set position window 17a tracking the preceding vehicle image 5a from the memory 8, and uses it as a reference image signal for calculating the inter-vehicle distance. Then, the microcomputer 10 selects an area corresponding to the window 17a in the memory 9 in which the image signal of the upper image sensor 4 is stored, and converts the image signal of the memory 9 pixel by pixel to the reference image signal. While sequentially shifting, the sum of the absolute values of the differences between the signals of the upper and lower pixels is calculated. That is, the position of the image that most closely matches the image within the window 17a is determined by sequentially shifting one pixel at a time. At this time, the area in the memory 9 involved in the calculation is the area 18 corresponding to the position of the window 17a, as shown in FIG. As mentioned above, the upper and lower pixels are compared, and the shift amount of the pixel when the sum of the absolute values of the difference signals is the minimum is a pixel, the pixel pitch is P, the base length of the optical system is L, and the lens 1 , 2, the distance R to the preceding vehicle is determined by the following equation.

[0017]

[Math 3]

In this way, the position of the window 17a that tracks the image of the preceding vehicle 5a is set with the image of the preceding vehicle 5a in the center, and stable image tracking becomes possible. Even if the preceding vehicle 5 moves left or right, it can be tracked and the inter-vehicle distance from the own vehicle can be accurately and continuously determined.

[0019]

As described above, according to the present invention, there is provided an image sensor for capturing an image of a preceding vehicle using substantially a pair of upper and lower optical systems, a display means for displaying an image of the preceding vehicle captured by this image sensor, A window surrounding the preceding vehicle image, an image tracking means for tracking the preceding vehicle image using this window, a binarization processing means for extracting edge components of the input image, and a binary image from which the edge components have been extracted are used. Since it is equipped with a window position correction means that determines the symmetry and corrects the position of the window, and a distance detection means that calculates the distance to the object surrounded by this window, the distance to the preceding vehicle can be accurately measured. can.

[0020] Furthermore, even when multiple preceding vehicles are running, the target preceding vehicle is indicated by a window, and the driver can easily know which preceding vehicle his or her vehicle is following and whose inter-vehicle distance is being detected. be able to.

Furthermore, since the symmetry is determined using the binary image from which edge components have been extracted and the window position is corrected, it is possible to prevent a positional shift between the preceding vehicle image and the window. , stable image tracking is now possible.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram of an inter-vehicle distance detection device according to an embodiment of the present invention.

FIG. 2 is a diagram of a display screen showing a state in which an image of a preceding vehicle has appeared on the display screen.

FIG. 3 is a diagram of a display screen showing a state in which an image of a preceding vehicle is within a window on the display screen.

FIG. 4 is a diagram of a display screen showing an image area to be subjected to binarization processing in an embodiment of the present invention.

FIG. 5 is a diagram showing a binary image in an embodiment of the present invention.

FIG. 6 is a diagram showing a state in which the set position of a window is corrected in an embodiment of the present invention.

FIG. 7 is a diagram showing an image area to be compared with a reference image for distance calculation in an embodiment of the present invention.

FIG. 8 is a configuration diagram showing a conventional distance detection device.

[Explanation of symbols]

1 Level 2 Lens 3 Image sensor 4 Image sensor 5 Leading vehicle 8 Memory 9 Memory 10 Microcomputer 11 Display screen (display means) 12 Image tracking device (image tracking means) 14 Binarization processing device (binarization processing means) 15 Window position correction device (window position correction means) 16 Memory 17 Window

Claims (1)

[Claims] Claim 1: The image signals formed on the image sensor by a pair of upper and lower optical systems are compared, and the deviation between the two images is electrically detected, based on the principle of triangulation. In the inter-vehicle distance detection device that measures the distance to the vehicle, the device includes a display means for displaying an image of the preceding vehicle captured by one image sensor, a window surrounding the image of the preceding vehicle, and an image tracking system for image-tracking the image of the preceding vehicle using the window. a binarization processing means for extracting edge components of an input image, and determining symmetry within a window using the binary image from which the edge components have been extracted, and positioning an image with good symmetry in the center thereof. A window position correction means corrects the position of the window surrounding the image of the preceding vehicle, and detects the deviation between the upper and lower corresponding image signals using the image signal inside the window as a reference signal, and calculates the inter-vehicle distance between the host vehicle and the preceding vehicle. What is claimed is: 1. An inter-vehicle distance detecting device comprising distance detecting means for calculating .