JPH056494A - Moving object measuring instrument and measured traffic flow picture processing device - Google Patents

Abstract

PURPOSE:To present a measured traffic flow picture processing device which has a simple constitution and a high measurement precision and can detect plural vehicles simultaneously existing on a picture. CONSTITUTION:The image of a vehicle is picked up by a TV camera 1 controlled by a TV camera control circuit 2, and this picture data is stored in a frame memory 3. The existence area of the vehicle along the vehicle movement route is preliminarily set in accordance with this picture data by an existence area setting circuit 9. Data indicating the vehicle is extracted from picture data in the existence area by a vertical line detecting circuit 10 and is continuously and periodically stored in a data time series storage circuit 12 in time series. The existence of the vehicle moving in the preliminarily set existence area is discriminated in a vehicle discriminating circuit 17 based on the data time series picture obtained by the data time series storage circuit 12, and the state variable of the discriminated vehicle is measured by a vehicle information measuring circuit.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving object measuring device for detecting various quantities of states of a moving object, and more particularly to a traffic volume, a speed, an inter-vehicle distance, a density, an occupancy rate, a congestion length in a car traffic flow.
The present invention relates to a traffic flow measurement image processing device that detects a state quantity such as a time required for passing a section.

[0002]

2. Description of the Related Art As one of moving object measuring devices, a traffic flow measuring image processing device for measuring vehicle traffic information using a TV camera has been developed. FIG. 9 is a block diagram showing the configuration of a conventional traffic flow measurement image processing apparatus described in Inoue et al .: “Vehicle dynamics measurement trial”, 17th Image Engineering Conference (1986), 16-13, P295-298. It is a figure. In the figure, 1 is an image pickup device, for example, a TV camera, 2 is a TV camera control circuit for controlling the TV camera 1, 3 is a frame memory for capturing a digitized image pickup signal, and 4 is a window setting circuit for setting a vehicle extraction region. 5, a vehicle detection circuit, 6 a vehicle traveling position detection circuit,
Reference numeral 7 is a vehicle movement amount detection circuit, and 8 is a vehicle information measurement circuit.

Next, the operation will be described. TV camera 1
Captures images of the traveling vehicle at arbitrary frame intervals according to the TV camera control circuit 2. The frame memory 3 takes in data obtained by converting the image pickup signal of the TV camera 1 into digital data.
The window setting circuit 4 sets a window on the screen, which is the size of the vehicle on the road. The vehicle detection circuit 5 detects the vehicle by extracting the characteristics of the vehicle such as horizontal lines and vertical lines from the area set by the window setting circuit 4. The vehicle traveling position detection circuit 6 detects the traveling position on the screen of the vehicle detected by the vehicle detection circuit 5, and the vehicle movement amount detection circuit 7 detects the traveling position and the traveling position of the front frame of the vehicle. By comparison, the amount of movement of the vehicle on the screen is detected and transmitted to the window setting circuit 4. The window setting circuit 4 moves the window according to the moving amount.
The above operation is repeated to track the vehicle. Tracking ends when the vehicle disappears from the screen, and the vehicle information measurement circuit 8
It measures the speed of the vehicle and the distance to the vehicle in front.

[0004]

Since the conventional moving object measuring device or traffic flow measuring image processing device is configured as described above, there is a problem that a tracking process between frames is required and the configuration becomes complicated. There is a point. Further, when the vehicle cannot be detected due to the influence of noise or other vehicles, there is a problem that the vehicle cannot be tracked and the measurement accuracy decreases. Furthermore, even if there are multiple vehicles on the screen, 1
There was a problem that only one vehicle could be tracked.

The present invention has been made in order to solve the above-mentioned problems, and does not require the correspondence of moving objects between frames for tracking, has a simple structure, and has a part of the frames. It is possible to provide a moving object measurement device or a traffic flow measurement image processing device that can extract a vehicle even if information is lost, has high measurement accuracy, and can detect a plurality of moving objects that are present on the screen at the same time. To aim.

[0006]

According to a first aspect of the present invention, there is provided a moving object measuring apparatus comprising: an image pickup section for picking up a moving image of a moving object; control means for controlling the image pickup section; And a frame memory for storing image data of
In the moving object measuring device that processes the image data, detects the moving object, and obtains information about the moving object, the presence area that presets the existence region of the moving object along the moving path of the moving object from the image data. Area setting means, extraction means for extracting data indicating a moving object from the image data in the existing area, and data for continuously storing the data indicating the moving object extracted by the extracting means in a time series at a constant cycle A sequence accumulating means is provided to detect a temporal change of a moving object moving in the preset existing area.

The moving object measuring device according to claim 2 of the present invention is
The data time series accumulation image obtained by the data time series accumulation means is binarized to create two-dimensional binary data 2
When the binarizing means and the data on both sides of the image data binarized by the binarizing means that extend in a predetermined direction of the reference binary data are the same binary data as the reference data, There is provided a padding means for changing the binary data in the direction perpendicular to the predetermined direction of the reference data to the same data as the reference data.

The moving object measuring apparatus according to claim 3 of the present invention is
The data time series accumulation image obtained by the data time series accumulation means is binarized to create two-dimensional binary data 2
In the predetermined direction of the reference binary data of the binarizing means and the image data binarized by the binarizing means,
When, on one side of the reference data, binary data having the same value as the reference data continues, and binary data having a value different from the reference data continues in the opposite direction to the predetermined direction of the reference data, A line segment element extracting means for extracting the reference data as a line segment element is provided.

According to a fourth aspect of the present invention, there is provided a traffic flow measurement image processing apparatus, which is a TV camera for imaging a vehicle, a TV camera control circuit for controlling the TV camera, and a frame memory for storing image data from the TV camera. A presence area of the vehicle along the movement route of the vehicle, an existence area setting means for presetting the existence data from the image data, and an extraction means for extracting data indicating the vehicle from the image data in the existence area;
The preset presence exists from the data time series accumulating means for accumulating the data indicating the vehicle extracted by the extracting means in a time series continuously at a constant cycle, and the data time series image obtained by the data time series accumulating means. A vehicle determination means for determining the presence of a vehicle moving in the area and a state quantity measuring means for measuring the state quantity of the vehicle determined by the vehicle determination means are provided.

According to a fifth aspect of the present invention, there is provided a traffic flow measurement image processing device, wherein the data time series accumulation image obtained by the data time series accumulation means is binarized and binarized means for creating two-dimensional binary data. In the image data binarized by the binarizing means, when the two adjacent data extending in a predetermined direction of the reference binary data are the same binary data as the reference data, the reference data Is provided with a filling means for changing the binary data in the direction perpendicular to the predetermined direction to the same data as the reference data.

According to a sixth aspect of the present invention, there is provided a traffic flow measurement image processing apparatus which binarizes the data time series accumulated image obtained by the data time series accumulating means to create two-dimensional binary data. Of the image data binarized by the binarizing means, binary data having the same value as the reference data is continuously provided on one side of the reference data in a predetermined direction of the reference binary data. And a value of 2 different from the reference data in the direction opposite to the predetermined direction of the reference data.
When the value data are continuous, the line segment element extraction means is provided for extracting the reference data as a line segment element.

According to a seventh aspect of the present invention, there is provided a traffic flow measurement image processing apparatus, a TV camera for picking up an image of a vehicle, a TV camera control circuit for controlling the TV camera, and a frame memory for storing image data from the TV camera. A presence area of the vehicle along the movement route of the vehicle, an existence area setting means for presetting the existence data from the image data, and an extraction means for extracting data indicating the vehicle from the image data in the existence area;
Vehicle data for extracting the data incidental to the data extracted by the extracting means and referring to the incidental information of the incidental information extracting means for retaining and extracting the incidental information, and selecting the data indicating the vehicle extracted by the extracting means. From the selection means, the data time series accumulation means for continuously accumulating the data obtained by the vehicle data selection means in a time series at a constant cycle, and the data time series image obtained by the data time series accumulation means, A vehicle determination means for determining the presence of a vehicle moving in the set existing area and a state quantity measuring means for measuring the state quantity of the vehicle determined by the vehicle determination means are provided.

A traffic flow measurement image processing apparatus according to claim 8 of the present invention is a TV camera for picking up an image of a vehicle, a TV camera control circuit for controlling the TV camera, and a frame memory for storing image data from the TV camera. A presence area of the vehicle along the movement route of the vehicle, an existence area setting means for presetting the existence data from the image data, and an extraction means for extracting data indicating the vehicle from the image data in the existence area;
Additional information extracting means for extracting and holding information incidental to the data extracted by the extracting means, and data time series accumulating means for accumulating the data indicating the vehicle extracted by the extracting means in a time series continuously at a constant cycle. And a vehicle determination means for determining the presence of a vehicle moving in the preset presence area from the data time-series image obtained by the data time-series storage means, the additional information extraction means, and the vehicle determination means. It is provided with a vehicle motion detecting means for detecting the behavior of the vehicle by referring to the obtained data, and a state quantity measuring means for measuring the state quantity of the vehicle judged by the vehicle judging means.

[0014]

The moving object measuring device and the traffic flow measuring image processing device according to claims 1 and 4 of the present invention set a region where a moving object exists along a moving path of the moving object, and move within the existing region. Data indicating an object is extracted for each arbitrary number of frames in a fixed cycle, and these extracted data are continuously accumulated in time series to create a data time series accumulated image.

The moving object measuring device and the traffic flow measuring image processing device according to claims 2 and 5 of the present invention are binarized data obtained by binarizing the data time series accumulated image according to claim 1 and claim 4. On the other hand, when the data on both sides extending in the predetermined direction of the reference binary data is the same binary data as the reference data, the binary data in the direction perpendicular to the predetermined direction of the reference data is the same as the reference data. Change into data.

The moving object measuring device and the traffic flow measuring image processing device according to claims 3 and 6 of the present invention are binarized data obtained by binarizing the data time series accumulated image according to claim 1 and claim 4. On the other hand, when the binary data of the same value is continuous in the predetermined direction of the reference binary data, and the binary data of the value different from the reference data is continuous in the opposite direction to the predetermined direction of the reference data, The reference data is extracted as a line segment element.

According to a seventh aspect of the present invention, there is provided a traffic flow measurement image processing apparatus which sets a region in which a moving object exists along a moving path of the moving object, and in the present region, data indicating a vehicle is displayed for each frame of a constant cycle. , And at the same time, extracts and retains information incidental to the data indicating the vehicle, and based on the incidental information, selects only the probable vehicle among the data indicating the vehicle. Then, the data is sequentially accumulated in time series to create a data time series accumulated image.

The traffic flow measurement image processing apparatus according to claim 8 of the present invention sets a region where a moving object exists along a moving path of the moving object, and in the present region, data indicating a vehicle is set for each frame of a constant cycle. At the same time, an arbitrary number of frames are extracted, and at the same time, information incidental to the data indicating the vehicle is extracted and held. Data indicating the vehicle is continuously accumulated in time series to create a data time series accumulated image, the vehicle is detected using this data time series accumulated image, and the information attached to the vehicle data is referred to for detection. The behavior of the vehicle that has been taken.

[0019]

【Example】

Example 1. 1 is a block diagram showing the configuration of a traffic flow measurement image processing apparatus according to a first embodiment of the present invention. In the figure, 9 is a presence area setting circuit, 10 is a vertical line detection circuit, and 1 is a vertical line detection circuit.
1 is a projection data creation circuit, 12 is a data time series storage circuit, 13 is a binarization circuit, 14 is a fill-in circuit,
Reference numeral 15 is a line segment element extraction circuit, 16 is a line segment extraction circuit, 17 is a vehicle determination circuit, and 18 is a vehicle information measurement circuit.

Next, the operation will be described. TV camera 1
Captures images of the traveling vehicle at arbitrary frame intervals according to the TV camera control circuit 2. The frame memory 3 takes in data obtained by converting the image pickup signal of the TV camera 1 into digital data.
This data is two-dimensional image data, and hereinafter, the horizontal direction is the x direction and the vertical direction is the y direction. The existence area setting circuit 9 retrieves from the frame memory 3 the data of the existence area of the vehicle along the preset movement route of the vehicle. Figure 2
Is an explanatory diagram showing an example of an existing area set by using the existing area setting circuit 9 on a screen of a road imaged by a TV camera, and 20 is an existing area. The vertical line detection circuit 10
A vertical line edge is extracted from the data in the existing area by using, for example, a vertical line detection filter. At this time, vertical line edge data in which the position where the vertical line edge exists is 1 and the other position is 0 is created. The projection data creation circuit 11 projects vertical line edge data in the x direction, and creates one-dimensional projection data in which the size in the x direction is 1 and the size in the y direction is the screen size. FIG. 3 is an explanatory diagram showing vertical line edge data and projection data. Reference numeral 30 shows vertical line edge data and 31 shows projection data. Projection data 31
Is created in one frame. The above process is repeated in a fixed cycle. The data time series accumulation circuit 12 accumulates the projection data 31 in time series by a predetermined number and creates two-dimensional data. The binarization circuit 13 converts this two-dimensional data into an appropriate threshold value Th0.
Binarize with. FIG. 4 is an explanatory diagram showing binary data,
The horizontal direction indicates the time (hereinafter, t direction), the vertical direction indicates the y direction, and 40a and 40b indicate the locus of the vehicle. In Figure 4,
The time is getting slower from left to right.

In the binary data, when a vertical edge cannot be extracted due to concealment of another vehicle, a hole 41 is formed. The fill-in circuit 14 fills in the binary data by the following processing. Binary data in the y direction from bottom to top,
Further, the t direction is sequentially accessed from the left or the right to select the reference data. When the condition that the reference data is 1 and two data adjacent to the reference data in the t direction are both 1 is satisfied, the data above the reference data in the y direction is set to 1. If the condition is other than the above, nothing is done. Do this for all data. FIG. 5 is an explanatory diagram of the hole filling method, in which 50 indicates a part of the binary data before the hole filling, and 51 indicates the binary data obtained as a result of the above hole filling. In the figure, the shaded area indicates binary data having a value of 1, and the other portions represent data having a value of 0.

The line segment element extraction circuit 15 performs the following process on the data which is the element of the line segment showing the trajectory of a part of the vehicle. The padded binary data is accessed in order from the top or bottom in the y direction and from the left or right in the t direction to select the reference data. In the t direction, the data having the value of 1 is continuously set to the left side of the reference data in the t direction, and the data having the value of 0 is set to the right side of the reference data. When the condition of being continuous is satisfied, or when the reference data is 1 and in the y direction,
When the condition that data having a preset value of 1 is continuous above the reference data and data having a preset value of 0 is continuous below the reference data is satisfied, The reference data is extracted as an element of the line segment. Do this for all data. By this processing, the contour line on the right side of the area 1 in binary data is extracted as an element of a line segment. FIG. 6 is an explanatory view of the line segment element extraction method, in which 60 indicates a part of the binary data, and 61 indicates one piece of data having a value of 1 on the left side of the reference data for the above line segment element extraction. , And the result of processing under the condition that one data having a value of 0 exists on the right side of the reference data. In the figure, a shaded area 60 indicates binary data having a value of 1, and a shaded area 61 indicates data extracted as a line segment.

The line segment extraction circuit 16 connects the elements of the line segment to extract the line segment. The vehicle determination circuit 17 makes a determination based on the length of the line segment and the position on the screen, and extracts the line segment that becomes the vehicle. The vehicle information measuring circuit 18 measures the speed of the vehicle, the distance between the vehicle and the preceding vehicle, using the three-dimensional position information of the start point and the end point of the line segment extracted as the vehicle.

In the above embodiment, the vertical line detection circuit is provided for processing the data in the existing area, but any processing for extracting the characteristics of the vehicle may be used.

Although the binarization circuit is provided after the data time series storage circuit, it may be provided between the projection creation circuit and the data time series storage circuit. Although the hole filling circuit is provided, it may be omitted if the vehicle information is not missing.
Further, in the padding circuit, although the binary data is accessed from the bottom in the y direction to the top, the binary data may be accessed from the top to the bottom. In that case, the reference data is 1 and in the t direction,
When the condition that the two data adjacent to the reference data are both 1 is satisfied, the data below the reference data in the y direction is set to 1.

In the line segment element extraction circuit, the reference data is 1
In the t direction, data having a value of 1 is continuously set to the left side of the reference data, and data having a value of 0 is set to the right side of the reference data. Alternatively, the reference data is 1, and in the y-direction, data having a value of 1 which is a preset number above the reference data is continuous and has a value of 0 which is a preset number below the reference data. We used the condition that the data are continuous, but the reference data is 1, and in the t direction,
The preset number of data with a value of 0 is continuous on the left side of the reference data, and the preset number of data with a value of 1 is continuous on the right side of the reference data, or the reference data is 1 , In the y-direction, a predetermined number of data having a value of 0 continues above the reference data,
In addition, a condition that a preset number of data having a value of 1 continues below the reference data may be used. In this case, in the binary data, the contour line on the left side of the region of 1 is a line segment. Is extracted as an element of.

Regarding the value of the binary data, the value of 1 may be replaced with 0 and the value of 0 may be replaced with 1, and the same effect as that of the above embodiment is obtained.

Example 2. FIG. 7 is a block diagram showing the configuration of the traffic flow measurement image processing apparatus according to the second embodiment of the present invention. In the figure, 70 is a horizontal position extraction circuit, and 71 is a selection projection data creation circuit.

Next, the operation will be described. TV camera 1
Captures images of the traveling vehicle at arbitrary frame intervals according to the TV camera control circuit 2. The frame memory 3 takes in data obtained by converting the image pickup signal of the TV camera 1 into digital data.
This data is two-dimensional image data, and hereinafter, the horizontal direction is the x direction and the vertical direction is the y direction. The existence area setting circuit 9 retrieves from the frame memory 3 the data of the existence area of the vehicle along the preset movement route of the vehicle. The vertical line detection circuit 10 extracts a vertical line edge from the data in the existing area using, for example, a vertical line detection filter.
At this time, vertical line edge data in which the position where the vertical line edge exists is 1 and the other position is 0 is created. For each horizontal line, the horizontal position extraction circuit 70 determines the average value (x centroid) of horizontal positions (x coordinates) of all vertical line edges existing in the horizontal line and the leftmost vertical line edge on the horizontal line. The horizontal position (leftmost point) and the horizontal position of the rightmost vertical line edge (rightmost point) are extracted, and this data is held. The selected projection data creation circuit 71 refers to the data of the horizontal position extraction circuit 70, and is within the specified range of the x-centroid of the horizontal position, or the distance between the leftmost point and the rightmost point has a threshold value Th1 or more. Select a horizontal line that satisfies the condition that, and project all vertical line edge data of the selected horizontal line in the x direction,
One-dimensional projection data in which the x-direction size is 1 and the y-direction size is the screen size is created by setting the projection value of the unselected horizontal line to 0. Horizontal line selection is
This is performed to extract only the data that is likely to be the vertical line edge data of the vehicle to be extracted. One projection data 31 is created in one frame. The above process is repeated in a fixed cycle. The data time series accumulation circuit 12 accumulates the projection data 31 in time series by a predetermined number and creates two-dimensional data. The binarization circuit 13 binarizes this two-dimensional data with a threshold value Th0. The hole filling circuit 14 is the first embodiment.
The binarized two-dimensional data is filled by the method shown in FIG. The line segment element extraction circuit 15 extracts data, which is an element of a line segment indicating a part of the locus of the vehicle, from the filled two-dimensional data by the method described in the first embodiment. The line segment extraction circuit 16 connects the elements of the line segment to extract the line segment. The vehicle determination circuit 17 makes a determination based on the length of the line segment and the position on the screen, and extracts the line segment that becomes the vehicle. Vehicle information measuring circuit 18
Uses the three-dimensional position information of the start point and the end point of the line segment extracted as the vehicle to measure the speed of the vehicle, the distance between the vehicle and the preceding vehicle, and the like.

Example 3. FIG. 8 is a block diagram showing the configuration of the traffic flow measurement image processing apparatus according to the third embodiment of the present invention. In the figure, 80 is a vehicle motion detection circuit.

Next, the operation will be described. TV camera 1
Captures images of the traveling vehicle at arbitrary frame intervals according to the TV camera control circuit 2. The frame memory 3 takes in data obtained by converting the image pickup signal of the TV camera 1 into digital data.
This data is two-dimensional image data, and hereinafter, the horizontal direction is the x direction and the vertical direction is the y direction. The existence area setting circuit 9 retrieves from the frame memory 3 the data of the existence area of the vehicle along the preset movement route of the vehicle. The vertical line detection circuit 10 extracts a vertical line edge from the data in the existing area by using, for example, a vertical line detection filter. At this time, vertical line edge data in which the position where the vertical line edge exists is 1 and the other position is 0 is created.
For each horizontal line, the horizontal position extraction circuit 70 determines the average value (x centroid) of horizontal positions (x coordinates) of all vertical line edges existing in the horizontal line and the leftmost vertical line edge on the horizontal line. The horizontal position (leftmost point) and the horizontal position of the rightmost vertical line edge (rightmost point) are extracted, and this data is held. The projection data creation circuit 11
Vertical line edge data is projected in the x direction, and one-dimensional projection data in which the size in the x direction is 1 and the size in the y direction is the screen size is created. The projection data 31 is
One is created for each frame. The above process is repeated in a fixed cycle. The data time series accumulation circuit 12 accumulates the projection data 31 in time series by a predetermined number and creates two-dimensional data. The binarization circuit 13 binarizes this two-dimensional data with a threshold value Th0. The padding circuit 14 fills the binarized two-dimensional data with the method described in the first embodiment. The line segment element extraction circuit 15 extracts data, which is an element of a line segment indicating a part of the locus of the vehicle, from the filled two-dimensional data by the method described in the first embodiment. The line segment extraction circuit 16 connects the elements of the line segment to extract the line segment. The vehicle determination circuit 17 makes a determination based on the length of the line segment and the position on the screen, and extracts the line segment that becomes the vehicle. The vehicle operation detection circuit 80 detects the behavior of the vehicle for the line segment determined to be the vehicle, using the information of the horizontal position extraction circuit 70. The method of detecting the behavior of the vehicle will be described below. The vehicle motion detection circuit 80 acquires the x centroid of each element of the line segment determined to be the vehicle from the horizontal position extraction circuit 70,
When the average value of the distance difference between the line segment and the median in the horizontal direction of the existence region of the horizontal line where the x center of gravity exists is less than or equal to the threshold value Th2, this vehicle is running near the center. To determine. On the contrary, the average value of the distance difference between the line segment and the median in the horizontal direction of the existence region of the horizontal line where the x center of gravity exists is equal to or larger than the threshold Th3, and the leftmost point and the leftmost point of the element. If the average value of the distance differences between the right points is less than or equal to the threshold value Th4, this vehicle is determined to be a vehicle traveling outside the existing area and is excluded. Further, the distance difference between the x centroid with respect to the start point of the line segment and the horizontal center value of the existing region of the horizontal line where the start point exists is a threshold Th5 or less, and the x centroid with respect to the end point of the line segment and the presence region If the difference in distance between the horizontal center value of the horizontal line where the end point exists and the horizontal center value is greater than or equal to the threshold value Th6, the vehicle tends to deviate from the existing region (such as lane change or obstacle avoidance). Traveling). The vehicle information measuring circuit 18 measures the speed of the vehicle, the distance between the vehicle and the preceding vehicle, using the three-dimensional position information of the start point and the end point of the line segment extracted as the vehicle.

In the above-described embodiment, it is determined that the vehicle tends to deviate from the existing area (abnormal running such as lane change or obstacle avoidance) due to the positional relationship of the x center of gravity with respect to the start point and the end point of the line segment. , The x center of gravity of each element of the line segment may be acquired, and the positional relationship between the straight line obtained by linearly approximating them and the existing region may be used for the determination.

[0033]

As described above, according to the first and fourth aspects of the present invention, there is no need for associating a moving object between frames for tracking, the structure is simple, and a part of the frames is used. Provide a moving object measuring device or a traffic flow measuring image processing device that has high measurement accuracy and can detect a plurality of moving objects that are present on the screen at the same time because the vehicle can be extracted even if the vehicle information is missing. can do.

Further, according to the second and fifth aspects of the present invention, the data created by the data time series accumulating means according to the first and fourth aspects are efficiently filled in, and the moving object is moved. Has the effect of increasing the detection accuracy of.

Further, in the inventions of claim 3 and claim 6 of the present invention, the data created by the data time series accumulating means of claim 1 or claim 4 or claim 2 or claim 5 is claimed. It is possible to efficiently extract line segment elements from the data filled in by the filling means 5 and speed up the processing.

The invention of claim 7 of the present invention has the effect of increasing the extraction accuracy of the vehicle.

Further, in the invention of claim 8 of the present invention, the behavior of the vehicle can be detected, so that there is an effect that the range of application of the traffic flow measurement image processing apparatus is expanded.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a traffic flow measurement image processing apparatus according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram showing an example of an existing area according to the first embodiment of the present invention.

FIG. 3 is an explanatory diagram showing vertical line edge data and projection data according to the first embodiment of the present invention.

FIG. 4 is an explanatory diagram showing binarized data of a data time series accumulated image according to the first embodiment of the present invention.

FIG. 5 is an explanatory diagram of a hole filling method according to the first embodiment of the present invention.

FIG. 6 is an explanatory diagram of a line segment element extraction method according to the first embodiment of the present invention.

FIG. 7 is a block diagram showing a configuration of a traffic flow measurement image processing apparatus according to a second embodiment of the present invention.

FIG. 8 is a block diagram showing a configuration of a traffic flow measurement image processing apparatus according to a third embodiment of the present invention.

FIG. 9 is a block diagram showing a configuration of a conventional traffic flow measurement image processing device.

[Explanation of symbols]

1 TV camera 2 TV camera control circuit 3 frame memory 8 Vehicle information measurement circuit 9 Existence area setting circuit 10 Vertical line detection circuit 11 Projection creation circuit 12 data time series storage circuit 13 Binarization circuit 14 hole filling circuit 15 Line segment element extraction circuit 16 line segment extraction circuit 17 Vehicle determination circuit 18 Vehicle information measurement circuit 70 Horizontal position extraction circuit 71 Selection projection data creation circuit 80 Vehicle motion detection circuit

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Akinori Seki             8-1-1 Tsukaguchihonmachi, Amagasaki-shi Mitsubishi Electric             Industrial Systems Research Institute Co., Ltd.

Claims (8)

[Claims] 1. An image processing unit for processing the image data, comprising: an image capturing unit for capturing a moving image of a moving object; a control unit for controlling the image capturing unit; and a frame memory for storing image data from the image capturing unit. Then, in the moving object measuring device that detects the moving object and obtains the information of the moving object, the existing area of the moving object along the moving path of the moving object, the existing area setting unit that presets from the image data, Extraction means for extracting data representing a moving object from the image data in the existing area, and data time series accumulating means for accumulating data representing the moving object extracted by the extracting means in a time series continuously at a constant cycle. A moving object measuring device, wherein the moving object measuring device is provided so as to detect a temporal change of a moving object moving in the preset existing area. 2. A binarizing unit for binarizing the data time-series accumulated image obtained by the data time-series accumulating unit and creating two-dimensional binary data, and an image binarized by the binarizing unit. Of the data, when the two adjacent data extending in a predetermined direction of the reference binary data are the same binary data as the reference data, the binary data in the direction perpendicular to the predetermined direction of the reference data The moving object measuring device according to claim 1, further comprising a filling unit that changes the same data as the reference data. 3. A binarizing unit for binarizing the data time-series accumulated image obtained by the data time-series accumulating unit and creating two-dimensional binary data, and an image binarized by the binarizing unit. Of the data, in the predetermined direction of the reference binary data, one side of the reference data is continuous with the binary data having the same value as the reference data, and the direction opposite to the predetermined direction of the reference data. 2. The moving object measurement according to claim 1, further comprising line segment element extracting means for extracting the reference data as an element of the line segment when binary data having a different value from the reference data continues. apparatus. 4. A TV camera for imaging a vehicle and the TV
A TV camera control circuit for controlling the camera, a frame memory for storing the image data from the TV camera, and a presence area setting means for presetting the presence area of the vehicle along the moving route of the vehicle from the image data. An extracting means for extracting data indicating a vehicle from the image data in the existing area; a data time series accumulating means for continuously accumulating the data indicating the vehicle extracted by the extracting means in a time series at a constant cycle; From the data time series image obtained by the data time series accumulation means, the vehicle determination means for determining the presence of the vehicle moving in the preset existence area, and the state quantity of the vehicle determined by the vehicle determination means are measured. An image processing apparatus for traffic flow measurement, which is provided with a state quantity measuring means for performing the above. 5. A binarizing unit for binarizing the data time-series accumulated image obtained by the data time-series accumulating unit and creating two-dimensional binary data, and an image binarized by the binarizing unit. Of the data, when the two adjacent data extending in a predetermined direction of the reference binary data are the same binary data as the reference data, the binary data in the direction perpendicular to the predetermined direction of the reference data 5. The traffic flow measurement image processing apparatus according to claim 4, further comprising filling means for changing the same data as the reference data. 6. A binarizing unit for binarizing the data time-series accumulated image obtained by the data time-series accumulating unit and creating two-dimensional binary data, and an image binarized by the binarizing unit. Of the data, in the predetermined direction of the reference binary data, one side of the reference data is continuous with the binary data having the same value as the reference data, and the direction opposite to the predetermined direction of the reference data. 5. A traffic flow measurement device according to claim 4, further comprising line segment element extraction means for extracting the reference data as an element of a line segment when binary data having a different value from the reference data continues. Image processing device. 7. A TV camera for imaging a vehicle and the TV
A TV camera control circuit for controlling the camera, a frame memory for storing the image data from the TV camera, and a presence area setting means for presetting the presence area of the vehicle along the moving route of the vehicle from the image data. An extracting means for extracting data indicating a vehicle from the image data in the existing area, an auxiliary information extracting means for extracting and holding information incidental to the data extracted by the extracting means, and an incidental information of the incidental information extracting means. Vehicle data selecting means for selecting data indicating the vehicle extracted by the extracting means with reference to information, and data time series accumulating means for accumulating data obtained by the vehicle data selecting means in a time series continuously at a constant cycle. And a vehicle for determining the presence of a vehicle moving in the preset presence area from the data time-series image obtained by the data time-series storage means. And judging means, the vehicle determining means comprising the state quantity measuring means for measuring the state quantity of the determined vehicle traffic flow measurement image processing apparatus. 8. A TV camera for imaging a vehicle and the TV
A TV camera control circuit for controlling the camera, a frame memory for storing the image data from the TV camera, and a presence area setting means for presetting the presence area of the vehicle along the moving route of the vehicle from the image data. An extracting means for extracting data indicating a vehicle from the image data in the existing area, an auxiliary information extracting means for extracting and holding information incidental to the data extracted by the extracting means, and a vehicle extracted by the extracting means. Data time series accumulation means for continuously accumulating data indicating the time series in a constant cycle, and a data time series image obtained by the data time series accumulation means of a vehicle moving in the preset existence area. Vehicle motion detection for detecting the behavior of the vehicle by referring to the vehicle determination means for determining the existence, the additional information extraction means, and the data obtained by the vehicle determination means Stage and said vehicle judgment means comprising the state quantity measuring means for measuring the state quantity of the determined vehicle traffic flow measurement image processing apparatus.