JP2931709B2 - Traveling vehicle detection device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a traveling vehicle detection apparatus for automatically extracting and detecting only a portion corresponding to a vehicle from an image of the traveling vehicle, for each vehicle.

[0002]

2. Description of the Related Art When processing image data obtained by observing a road from the side or obliquely above the road, the image of the detection target vehicle is the same as or opposite to the background image and the detection target vehicle. The vehicle traveling in the traveling direction is stored as an image at the same time while being seen and hidden. Generally, the background image data changes with the passage of time due to factors such as the weather. Therefore, conventionally, the background image is automatically updated by checking the change in the image data of the same area over time for the entire image area. Had gone.

In such a case, the amount of calculation for detecting the change is large because the calculation is performed for the entire area. Further, when a plurality of vehicles are present in the image at the same time, it is not possible to detect where in the image area the background and where each vehicle is. Or, the vehicle detection is performed by a method such as taking a logical sum with respect to an image in which the time lapse is continuous only in the portion of (positive).

In such a conventional method, it is difficult to separate / detect a plurality of invisible vehicles by judging them as different vehicle regions without detecting a change in the entire screen region. Met.

[0005]

Conventionally, when processing background brightness changes due to factors such as weather changes over time, and image data in which a plurality of vehicles appear and disappear in an image at the same time, an image area is conventionally used. Large-scale calculations were required because changes over time were examined for the entire part, and it was difficult to automatically and stably determine the background and individual vehicle areas to separate / detect them. .

SUMMARY OF THE INVENTION It is an object of the present invention to stably separate / detect a background or an individual vehicle even when the brightness of the background changes due to a weather change or when a plurality of vehicles appear and disappear in an image. And

[0007]

According to the present invention, in order to solve the conventional problems, a preset image area is used.
The change with time is detected only for the pixel data of only the detection slit portion along the vehicle traveling direction. Furthermore, only the data of the detection slit part is accumulated with the passage of time, and the line segment is detected for the changed part to determine which part of the image area is the background or which vehicle and the amount of movement on the image. Each image is superimposed as a separate image for each vehicle, and can be separated and extracted for each vehicle.

That is, a means for sequentially accumulating pixel data of a preset slit portion of an image frame of a photographed image on a memory for a slit image, a means for detecting a line segment of the slit image, and a means for detecting the line segment. A circuit for obtaining a difference image from a background image recorded in advance for the obtained vehicle presence area, and a configuration in which the difference image is shifted and superimposed by the amount of vehicle movement obtained by the line segment detection.

[0009]

In the prior art, for all the pixel values of the photographed image, the logical sum of the difference image with the background is calculated for each frame to detect a vehicle as a moving object. In the present invention, however, the amount of calculation is reduced. In order to achieve this, it is noted that the moving direction of the vehicle is substantially constant, and the vehicle detection process is performed by calculating only a portion (slit portion) of the captured image (image frame) through which the vehicle passes. I have to.

In the present invention, a detection slit is set in advance in a digital image of a traveling vehicle along the traveling direction of the vehicle. In the case of photographing obliquely from above, a plurality of lines are set so that the vehicle travels within the slit. Various patterns can be considered as the setting of the detection slit, but as a simple example, a TV camera is installed on the side in the vehicle running direction and shooting is performed. The detection slit may be used as it is.

[0011]

FIG. 1 shows an embodiment of the present invention. In FIG. 1, reference numeral 1 denotes an image memory for storing a digital image input from a television camera, and reference numeral 2 denotes only a predetermined one or a plurality of detection slit portions. A pixel readout circuit, 3 is a storage memory for storing the detected slit data read out by the readout circuit 2 with time, 4 is a change detection circuit for data read out from the same image area due to the passage of time, and 5 is a background. A storage / update circuit 6 for storing / updating a slit image alone, a line segment detection for a changed portion detected by the change detection circuit 4 to detect a vehicle area, and a line segment detection for calculating a moving amount of the vehicle. Circuit 7,
A background image memory for storing / updating for each image area determined to be a background, 8 is a subtraction circuit for subtracting a background image from an input image, 9 is a difference image memory for storing a subtraction result, 1
Numeral 0 denotes a shift circuit for taking out individual vehicle areas determined by the change detection and line segment detection circuit 6 and shifting them by the vehicle movement amount. Numeral 11 denotes a vehicle memory for superimposing and storing these vehicle area images for each vehicle. Represents a circuit.

As a simple example, a case will be described in which a television camera is installed on the side of the vehicle running direction and the vehicle is observed so that the scanning direction and the vehicle running direction are parallel. As time passes, the vehicle starts to move from the right or left of the image to the image area according to the traveling direction, passes through the center of the image, and disappears in the opposite direction.

The one-dimensional data is accumulated with the passage of time, and the scanning line direction in the original image area is set in the sub-scanning direction. The image is stored as two-dimensional data in the main scanning direction, and this image is hereinafter referred to as a spatiotemporal image. Considering the change with time at the same position of the original image, that is, the same position of the scanning line set in the slit, that is, the change in data in the direction of the main scanning in the spatiotemporal image, Although there is almost no change in the background, a change occurs in the vehicle portion as a result of the corresponding point moving in the detection slit direction as the vehicle travels. Therefore, a pixel having a large change with time at the same position of the scanning line set in the slit can be detected as a traveling vehicle.

When the detected changed pixel is viewed in a spatiotemporal image, a locus on a line corresponding to a running mode such as speed, acceleration, and moving direction is drawn. FIG. 2 shows information appearing in the spatiotemporal image.

Reference numerals 20-1, 20-2, 20- in the figure
3, ... are multiple images at the same position captured by the camera,
.. Represent the images on the scanning lines set in the slits, and 22 represents the spatiotemporal image. The spatiotemporal image 22 includes an image 21-1 on the image 20-1 and an image 20-
2 on image 2 and image 21-3 on image 20-3
, ... are stored in chronological order. As a result, on the spatiotemporal image 22, a locus descending to the right appears for a vehicle traveling from left to right, a locus descending to the left appears for a vehicle traveling from right to left, and a vehicle traveling from left to right appears. When the vehicle accelerates, the slope of the rightward trajectory gradually changes.

If line segments are detected for a locus in the spatiotemporal image, it is possible to determine which region of the image is the background and which region the vehicle is at at each time, and the amount of pixel movement in the traveling direction. Even if a plurality of vehicles exist in the image area at the same time, it is possible to detect the number of vehicles by dividing the image in the slit direction and extracting line segments by a known processing technique.

The background area is B, and the area including the background and each of the i vehicles appearing in the image at the same time is F ₀ ,
Assuming that F ₁ ,..., F _n , regions of the same type should be arranged in the traveling direction, that is, the direction orthogonal to the detection slit. FIG. 3 shows an example of a vehicle traveling in the opposite direction passing by in an image area being photographed immediately before overlapping, and FIG. 4 shows an example in which the vehicle crosses and overlaps, thereby causing visibility. Therefore, the background area in the original image can be known as it is from the area regarded as the background in the detection slit. Therefore,
The background image stored so far can be automatically updated for the area regarded as the background.

A so-called difference image can be created by subtracting the background image from the original image. The image region in which only individual vehicles exist can be determined by the above spatiotemporal image processing. If only the same vehicle portion is superimposed in the traveling direction by the amount of movement accompanying time change, the individual vehicles can be separated. / Can be detected.

The brightness or color information of the image data is stored as data expressed by two-dimensional coordinates of the vehicle traveling direction (ie, the scanning line direction of the image data) x and two axes orthogonal to the x and the observation time t. And this value is changed to f (x,
y, t) and accumulate sequentially over time. Here, the image data area stored at each time is defined as 0 ≦ x ≦ (X−1) and 0 ≦ y ≦ (Y−1).

When the height near the bumper of an ordinary vehicle is y = y _slit in the image, the _slit is set to y = y _slit . In this case, as the slit images at each time, f (0, y _slit , t) and f ((X−1),
_Read out one scan line data having both ends (y _slit , t).
In the case of this example, the readout circuit 2 sets the y coordinate of the pixel to y at the time when the image data is stored in the image memory 1 at each time.
_It is determined whether or not it matches the _slit, and only when it matches, the data is output to the storage memory 3. The data read from the slit is stored in the storage memory 3 as data expressed by the order u in the order of reading along the slit and the time t, and this value is represented by s (u, t). The information is sequentially stored as the time elapses.

In this example, the _slit is set to y = y _slit , and pixel f (0, y _slit , t) is _shifted to pixel f ((X−1),
y _slit, t) to so read the order becomes a u≡x. The data read from the slit is considered as two-dimensional data having u and t as two axes, and is hereinafter referred to as a slit image.

In the t-axis direction of the slit image, pixels having the same coordinates are arranged in order in the original image area as time elapses. Therefore, if a background in which the pixel does not move is stored, the change in the t direction is small due to only a gradual change caused by a weather change or the like. This value is defined as sb (u). If the vehicle crosses the image area over time,
If a vehicle portion appears at that pixel at a certain time t = t _in and another background shifts again to the pixel at t = t _out , the value at t _in ≦ t ≦ t _out is as large as sb (u). Differences arise. FIG. 5 shows a change example of the slit image.

Therefore, the change detection circuit 4 checks the change in the slit image in the t-axis direction, and at time t, it can be determined whether each area in the slit image is a background or a vehicle. Each pixel s (u, t) of the slit image determined as the background is stored / updated by the circuit 5 as a background slit image. At the same time, in the original image area, all pixels in the direction orthogonal to the slit determined to be background in the original image area, in the example, the horizontal direction because the slit is horizontal, are determined as background, and all the area is stored as a background image. Stored in circuit 7 /
Update.

If a change is detected in the slit image by detecting a change, the vehicle has crossed the image area. Therefore, as shown in FIG. 5, a line locus corresponding to the speed of the vehicle is drawn. If the velocity is constant, the curve becomes a linear trajectory. If the vehicle is accelerating, the curve is curved in the positive direction of time. If the vehicle is decelerated, the curve is a curve curved in the negative direction of time. Therefore, if the line segment is detected, the speed of the vehicle on the image, that is, the moving pixel amount on the image area at a unit time change can be detected.

To detect a line segment in the slit image, a known method may be used. For example, linear regression by the least squares error method is performed on slit areas for several adjacent frames.

In the case where vehicles running in opposite directions just intersect in the image area being photographed and two or more vehicles appear and disappear in the stored image, the line segment detection circuit 6 determines the number of vehicles according to the number of vehicles. Since only the number can be detected,
A region where only one vehicle appears can be determined, and only a background or a region where two or more vehicles appear can be deleted. FIG. 6 shows an example of determination and deletion. At time t, the background image is subtracted from the image acquired by the television camera by the circuit 8 and the difference image is stored in the memory 9. When the circuit 10 is superimposed in the traveling direction according to the moving amount of each vehicle, Identical parts can be aligned and overlapped. Zones surrounded by dotted lines in the figure are erased zones. FIG. 7 shows an example of superposition using the movement amounts of the individual vehicles detected from the slit image. According to the above method, the images of the individual vehicles can be obtained separately and stored in the vehicle storage circuit 11.

As a simple example, the case where the photographing is performed from the side has been described. However, the present invention is effective even when the photographing is performed obliquely from above. For example, in the case of FIG. 8, two or more slits may be set along the running direction, and data read from that area may be combined.

That is, as shown in FIG. 9, images 23-1, 23-2, and 23 corresponding to a plurality of scanning lines on image 20-i.
-3 is extracted, and the images 23-1, 23-2, and 23-3 are extracted.
May be arranged on one horizontal scanning line to create a spatiotemporal image 22.

[0029]

As described above, according to the present invention, by detecting a change with time in only a preset detection slit portion of the entire image area and detecting a line segment for the changed portion, Detects whether each area of the image data is a background or a vehicle. Even if the brightness changes due to weather changes, or if multiple vehicles appear or disappear in the image at the same time, a smaller scale It is possible to provide a traveling vehicle detection device operable only by calculation.

[Brief description of the drawings]

FIG. 1 shows an embodiment of the present invention.

FIG. 2 is a diagram illustrating information appearing in a spatiotemporal image.

FIG. 3 shows a state before and after the vehicles overlap.

FIG. 4 shows a state in which vehicles overlap.

FIG. 5 is a diagram illustrating background detection by detecting a change in a slit image.

FIG. 6 is a diagram illustrating the determination of a vehicle appearance region and the elimination of an overlapping region.

FIG. 7 is a diagram illustrating individual superposition of vehicle images using a vehicle movement amount.

FIG. 8 shows an example in which a plurality of slits are set.

FIG. 9 shows an example of a case where an image is taken from obliquely above.

[Explanation of symbols]

 Reference Signs List 1 image memory 2 pixel readout circuit 3 storage memory 4 change detection circuit 5 storage / update circuit 6 line segment detection circuit 7 background image memory 8 subtraction circuit 9 difference image memory 10 shift circuit 11 vehicle storage circuit 20 image 21 image on scan line 22 Spatio-temporal image

──────────────────────────────────────────────────の Continued on the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) B60R 21/00 621 B60R 21/00 624 G06F 15/62 380-420 G06F 15/66-15/68 G06F 15 / 70-15/70 465 H04N 7/18

Claims (1)

(57) [Claims] 1. An image memory circuit for storing a digital image input from a television camera, and one or a plurality of image memories arranged in an image area stored in the image memory and arranged in a vehicle traveling direction on the image. A circuit for reading out pixel values of only the detection slit portion of the book, a memory circuit for sequentially storing pixel data at the same time read out by the readout circuit, and a memory circuit for reading data from the same image area inside the memory circuit. A circuit for detecting a change with time of the output data, and a storage / update for storing / updating a slit image consisting only of the background based on the output of the change detection circuit.
An update circuit, a circuit that detects a line segment in a space having a coordinate axis orthogonal to the read image area coordinates and time with respect to the change detected by the change detection circuit, and a background by a change detection and line segment detection circuit. A background image memory circuit for storing / updating for each image area corresponding to the in-slit area determined as above, a circuit for calculating and outputting the difference between the background image memory circuit and the image memory circuit for storing the input image, and storing the difference image A circuit for obtaining a difference image based on the vehicle area on the image and the vehicle movement amount on the image corresponding to the slit determined to be the vehicle area by the change detection and line segment detection circuits; and a circuit for superimposing the difference image. A traveling vehicle detection device characterized by comprising: