JPH06325179A - Automatic speed measuring instrument - Google Patents

Abstract

PURPOSE:To disable a state wherein a speed is being measured to be detected in advance and to flexibly select a measurement place by tracking a vehicle in a picked-up image and measuring the speed from the number of frames wherein the body passes a range whose distance is already known. CONSTITUTION:A motion area extracting process part 2 extracts a motion area from an image signal generated by picking up an image through a image pickup part 1. Further, a motion area feature point calculating process part 3 performs a process for finding feature points of the motion area found by the motion area extracting process part 2. Then a prescribed range passage deciding process part 4 measures the number of frames required to pass the range whose distance is already known from the feature points (center of gravity in this case) found by the motion area feature point calculating process part 3. Namely, the absolute coordinate values of the start point and end point of speed measurement which are found by a known point coordinate input process part 5 are stored and the time required for the center of gravity of the area of the vehicle to pass between specific points is measured.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic speed measuring device for automatically measuring the speed of a moving body using a camera.

[0002]

2. Description of the Related Art Conventionally, in an automatic speed measuring device such as an automobile, a vehicle under measurement is irradiated with a radio wave such as a microwave and the frequency deviation of a reflected wave is measured. The reflected frequency can be determined by the following equation (1) due to the Doppler effect, and from this the speed of the vehicle under test can be measured. Where f is the frequency of the microwave to be emitted, f'is the frequency of the reflected wave obtained by the measurement, v is the speed of the vehicle under measurement,
c is the speed of light.

F '= f * c / (cv) (1) FIG. 16 shows a schematic diagram of a conventional automatic speed measuring system. In the figure, 100 is a reflected wave detection device, 101 is a microwave emission device, 102 is a traveling vehicle, and 103 is a one-lane road on one side. When measuring the speed on the one-lane road 103 on one side,
First, the microwave emission device 101 emits a microwave of frequency f toward the traveling vehicle 102, the reflected wave thereof is detected by the reflected wave detection device 100, and the frequency of the detected reflected wave is defined as f ′. From this, the vehicle speed v can be obtained from the above equation (1).

Since this conventional method is a method of measuring the speed by irradiating the measured vehicle with microwaves, when the measured vehicle is equipped with a detector for detecting microwaves,
It is possible to detect in advance that speed measurement is in progress, and there is a problem that the control becomes difficult, especially when a vehicle speed violation is to be controlled.

Further, since this conventional method is a method for measuring the reflected wave from the vehicle to be measured, it is possible to measure the speed of the leading vehicle when a plurality of vehicles enter the measurement area in series. However, after that, there was a problem that it was difficult to identify which vehicle was the reflected wave.

Further, in such conventional measurement, the microwave emission device and the reflected wave detection device must be installed on the extension of the traveling direction of the vehicle to be measured, and the reflected wave must be measured. There is a problem that it is difficult to measure in parallel with the traveling direction of, and in this prior art, since the velocity is obtained from the reflected wave of the microwave, the instantaneous velocity is measured. There was a problem that the average speed of a certain section of the road could not be obtained. Due to these two problems, there is a problem that the installation location of the speed measuring device is limited.

Further, when measuring the velocity of another animal body by this conventional method, since the reflected wave of the microwave is used, it is necessary that the outer skin of the velocity-measured animal body be metal. There is a problem that it is difficult to measure the speed of people and people.

[0008]

SUMMARY OF THE INVENTION According to the present invention, since it is possible to detect in advance that a vehicle to be measured is in a speed measurement state by detecting a microwave, it is difficult to control the speed. An object of the present invention is to provide an automatic speed measuring device which solves the problem that only the leading vehicle can measure the speed when a series of vehicles enter.

The present invention solves the problems that it is difficult to measure parallel to the traveling direction of the vehicle to be measured and that the average speed of the vehicle to be measured cannot be obtained in the non-linear measurement section. However, the object is to provide a flexible automatic speed measuring device at the installation location.

Another object of the present invention is to provide an automatic speed measuring device capable of measuring the speed even when the object to be measured has a skin other than metal.

[0011]

In order to achieve the above-mentioned object, the present invention provides an automatic speed measuring device for automatically measuring the speed of a moving animal body in an image from an image signal input from an image input device. A means for extracting a region of the body, a means for calculating the feature points of the extracted region of the same moving body for each frame or field while tracking the moving body, and velocity measurement in which the distance of the section for measuring the velocity is known. A means for inputting the start point and the velocity measurement end point as coordinate values on the image, and the calculated region feature point coordinate on the image of the moving object passes between the velocity measurement start point coordinate and the velocity measurement end point coordinate. There are provided means for calculating the number of frames or fields required for the calculation, and means for calculating the speed of the moving body from the calculated number of frames or fields and the distance of the section for measuring the speed.

Further, when extracting the region of the moving object, a means for accumulating a background image signal in a memory in the absence of the moving object in advance, and the accumulated background signal and the input image signal in which the moving object exists The means for calculating the difference and the means for binarizing the calculated difference signal by a preset threshold value are used.

In order to effectively track the moving object, the center of gravity of the area is used as a feature point of the moving object area to track the moving object.

[0014]

According to the present invention, a closed region of a vehicle is extracted by image processing from an image signal picked up by an image input device such as a television camera, a center of gravity of the region is obtained, and a region whose distance is known in advance is set as the center of gravity. The main feature is that the speed of the vehicle is measured based on the time that the vehicle passes. The present invention is a technique for finding a feature point from a captured image signal by image processing to track the moving object and finding the speed from the time when the feature point passes a known distance. The technique differs from the technique of actively transmitting microwaves from the measuring device and obtaining the velocity from the reflected wave in that the velocity can be passively measured without actively transmitting microwaves from the measuring device side.

[0015]

FIG. 1 is a diagram for explaining an embodiment of the present invention,
Reference numeral 1 is an image pickup unit, 2 is a moving region extraction processing unit, 3 is a moving region feature point calculation processing unit, 4 is a predetermined range passage determination processing unit, 5 is a known point coordinate input processing unit, and 6 is a speed calculation processing unit. First, specify the absolute coordinate system of the image to be handled. FIG. 2 is a diagram showing a coordinate system of an image. For example, the NTSC television signal is 1
When sampling at 4.3 MHz, there are 910 pixels in the horizontal direction and 525 lines in the vertical direction. However, considering effective pixels, the horizontal direction is 768 and the vertical direction is 480 as shown in FIG.
Becomes In this embodiment, this is the absolute coordinate.

A moving area is extracted from an image signal picked up by the image pickup section 1 by a moving area extraction processing section 2. Although there are many methods of extracting a moving area, an example of a method of calculating a difference between a background image signal and an input image signal to obtain a moving area will be described here.

FIG. 3 is a diagram for explaining in detail an example of the moving area extraction processing unit 2 of FIG. In the figure, 20 is a background image memory unit, 21 is a difference processing unit, and 22 is a binarization processing unit. The background image (image in which no vehicle exists) captured by the image capturing unit 1 is stored in the background image memory unit 20 in advance. Next, the image pickup unit 1 restarts the image pickup in the state where the vehicle exists, and the difference processing unit 21 calculates the absolute value difference from the background image. The signal for which the absolute value difference has been calculated is input to the binarization processing unit 22, binarized with an appropriate threshold value, and input to the moving region feature point calculation processing unit 3.

The moving area feature point calculation processing section 3 performs processing for finding the moving area feature points found by the moving area extraction processing section 2. As a feature point of the moving area, the center of gravity of the area or the maximum x
(Or y) coordinates can be used, but
Hereinafter, an example of a method of obtaining the center of gravity using the labeling process will be described.

In the labeling processing, since the same label number is given to the same closed region, the region can be identified by the label number. However, in the conventional labeling process, the label number given to each processing frame may differ depending on the positional relationship of the closed region. An example in which the label number assigned to each processing frame differs depending on the positional relationship will be described below.

FIG. 4 shows an example of a binary image of a road with three lanes on each side taken from the front in the traveling direction. In the figure, 60 is an area of a car A, 61 is an area of a car B, 62 is an area of a car C, 63 is a speed measurement start point s, 64 is a speed measurement end point e, 65 is a three-lane road on one side, and 66 is an image. Represents an image frame. Here, the distance between the speed measurement start point s and the speed measurement end point e is y (m).

In normal labeling, label numbers are provisionally provided sequentially from the upper left of the image in the scanning line direction, and the same label number is re-assigned to the areas finally determined to be the same area. Therefore, in the case of the example shown in FIG. 4, the label number 1 is assigned to the area 60 of the vehicle A, the label number 2 is assigned to the area 61 of the vehicle B, and the label number 3 is assigned to the area 62 of the vehicle C. However, since each car runs at different speeds in the traveling direction, the positional relationship may change during the speed measurement period.

FIG. 5 shows an example of a similar binary image after several to several tens of frames from the image of FIG. In this case, in the conventional labeling process, the label number 3 is assigned to the area 60 of the vehicle A, the label number 1 is assigned to the area 61 of the vehicle B, and the label number 2 is assigned to the area 62 of the vehicle C. Therefore, tracking with the same label number becomes difficult.

Therefore, as a tracking method using the labeling process, the technique proposed in the moving object detection and tracking processing method of Japanese Patent Application No. 3-227627 and the moving object tracking processing method of Japanese Patent Application No. 4-29145 is proposed. A car with the same label number is used to track and the center of gravity of the car is calculated.

In the moving object detection and tracking processing method proposed in Japanese Patent Application No. 3-227627, the difference between the background image and the input image is calculated for each frame, and the difference image of the frame is effectively moved at an appropriate threshold value. 2 divided into area and background area
The initial value of the cutout center point is calculated by detecting the effective moving area from the first frame image, and the cutout center point or the center of gravity obtained by the labeling process of the previous frame image is calculated from the next frame image. A rectangular frame is cut out based on the cut-out center point calculated from, and the labeling process is performed in the rectangular frame area starting from the cut-out center point, and the label number and the center of gravity point of the moving area are obtained and sequentially obtained for each frame. The input image is cut out based on the cut out center point, and the moving object is recognized.

Further, according to the moving object tracking processing method proposed in Japanese Patent Application No. 4-29145, when the labeling processing is performed in the rectangular frame including the closed moving object region cut out from the inter-frame difference block image, Starting from the center of gravity obtained in the image, the labeling process is performed by changing the scanning direction such as the presence or absence of the label number in a spiral shape, and a new moving object enters the rectangular frame for labeling process. Even so, the tracking process can be performed correctly.

An example using these techniques will be described below. FIG. 6 is a diagram for explaining an example in which the moving object tracking processing method proposed in Japanese Patent Application No. 3-227627 and Japanese Patent Application No. 4-29145 is applied to the image shown in FIG. In the figure, 70 is a labeling local area a, 71 is a labeling local area b, 72 is a labeling local area c, and 73 is a center of gravity WA (WAx, WAY) of the vehicle A.

In the local area, the center of gravity obtained by the labeling process one frame before is used as the starting point, and the spiral labeling process is performed on the next processing frame. On a road with three lanes on each side, the labeling process is performed in parallel in three times, so that the spiral labeling process of three local areas is performed at the same time. The center of gravity of the area of label number 1 obtained in each local area is calculated for each frame, and the calculated center of gravity is set as the starting point of the spiral labeling processing in the next processing frame. In the following description, the process for the area 60 of the vehicle A will be described.
1, the process for the area 62 of the car C
The process is the same as the process in. With this method, it is possible to solve the problem that the label number differs depending on the positional relationship of the closed region. The calculated center of gravity is input to the predetermined range passage determination processing unit 4.

In the ordinary labeling process, the label numbers are tentatively given and then the concatenated label numbers are integrated into one real label number, but this process requires a huge amount of processing. Therefore, real-time processing was difficult. In this automatic velocity measuring device, since the labeling process is performed for each frame to calculate the center of gravity, the real time process (within 1/30 seconds) is indispensable. As a method for enabling real-time processing, for example, Japanese Patent Application No. 14782/1992.
It is conceivable to use the technology of the image processing apparatus proposed in No. In this image processing apparatus, when a connection relation is detected when a label number is provisionally assigned, the connection relation of the label is stored and the assigned label number is CA.
It is stored in M (Content Addressable Memory). At the time of this storage, of the provisional label numbers in the stored connection relation, for example, the smallest provisional label number is used as a representative value as a real label number based on the connection relation, and other provisional label numbers are connected to the representative label number. The tentative label number is retrieved from the CAM, and all the retrieved tentative label numbers are rewritten with the representative label number at the same time.

By using such a technique, that is, a technique that enables real-time processing using CAM, it is possible to realize a practical speed measuring process even in an automatic speed measuring device using a labeling process. .

Next, the known point coordinate input processing section 5 will be described. The known point coordinate input processing unit 5 inputs the coordinate values of the speed measurement start point s63 and the speed measurement end point e64 which are set in advance. Although many input methods are conceivable, an example using a pen type light receiving element will be described here.

FIG. 7 shows an example of a coordinate value input device using a pen type light receiving element. In the figure, 50 is a display, 5
Reference numeral 1 is a pen type light receiving element, and 52 is a coordinate value calculation processing unit.
The input image is displayed on the display display 50.
Therefore, the point on the image to be input, that is, the speed measurement start point s63 and the speed measurement end point e64 are pressed by the pen-shaped light receiving element 51, respectively. When the pen-shaped light receiving element 51 detects the light emission of the scanning line which scans the pressed point during the pressed time, the pen-shaped light receiving element 51 outputs the detection information at the timing.
Enter 2. The coordinate value calculation processing unit 52 calculates the elapsed time of the horizontal scanning lines and the number of lines of the vertical scanning lines which coincide with the time when the detection information is received, and calculates the elapsed time of the horizontal scanning lines and the calculated vertical scanning line. The absolute coordinates at the position where the pen type light receiving element 51 is pressed are calculated from the number of lines.

The absolute coordinate values s (sx, sy) and e (ex, ey) of the speed measurement start point s63 and the speed measurement end point e64 thus obtained are stored in the predetermined range passage determination processing unit 4. Is entered. Note that the known point coordinate input processing in the known point coordinate input processing unit 5 does not need to be input again unless the measurement position is changed, if it is input once before the speed measurement.

The predetermined range passage determination processing unit 4 passes a known range of distances in advance from the feature point (the center of gravity point in this embodiment) obtained by the moving region feature point calculation processing unit 3. Measure the number of frames required. That is,
The absolute coordinate values s (sx, sy), e (ex, ey) of the speed measurement start point s63 and the speed measurement end point e64 obtained by the known point coordinate input processing unit 5 are stored in advance, Center of gravity point WA (WAx, Way) 7 of area 60
The time required for 3 to pass between se is measured.

FIG. 8 shows the flow of processing in the predetermined range passage determination processing unit 4. In this example, since the velocity in the y direction is measured as can be seen from FIG.
Use only coordinates. The flow of this processing will be described below according to the processing (a) to (h) shown in FIG.

(A) The frame number counter FN is set to 1. (b) Input the y coordinate WAy of the center of gravity of the area 60 of the car A. (c) If WAy is greater than or equal to the y coordinate sy of the speed measurement start point, proceed to processing (d). In other cases, transition to the next processing frame image,
In the processing (b) state, the input of the y-coordinate WAy of the center of gravity point of the next processing frame image is waited for.

(D) Transition to the next processing frame image. (e) Wait for the input of WAY. (f) The frame number counter FN is incremented by 1.

(G) WAy is the y coordinate ey of the speed measurement end point
If it exceeds, proceed to processing (h). Otherwise, proceed to process (d). (h) Input the value of the frame number counter FN into the distance frame number K.

In the above embodiment, y as shown in FIG. 6 is used.
Although the case of the road parallel to the direction has been described, the following method is used to measure the speed of the vehicle traveling straight in the direction parallel to the x direction. FIG. 9 is a diagram showing an example in which a road with three lanes on each side is imaged from a direction perpendicular to the traveling direction of the vehicle. In the figure, 80 is the region of the car D, 81 is the local region d for labeling, and 82 is the center of gravity WD (WDx, WD) of the car D.
represents y).

When the speed is measured from the direction perpendicular to the road as shown in FIG. 9, since the speed in the x direction is measured, only the x coordinate is used for speed measurement in speed calculation. . FIG. 10 shows the flow of processing in the example of FIG.
The flow of this processing will be described below according to the processing (a) to (h) shown in FIG.

(A) The frame number counter FN is set to 1. (b) Input the x coordinate WDx of the center of gravity of the area 80 of the car D. (c) If WDx is greater than or equal to the x coordinate sx of the speed measurement start point, proceed to processing (d). In other cases, transition to the next processing frame image,
In the processing (b), the input of the x coordinate WDx of the center of gravity point of the next processing frame image is waited for.

(D) Transition to the next processing frame image. (e) Wait for WDx input. (f) The frame number counter FN is incremented by 1.

(G) WDx is the x coordinate ex of the speed measurement end point
If it exceeds, proceed to processing (h). Otherwise, proceed to process (d). (h) Input the value of the frame number counter FN into the distance frame number K.

FIG. 11 is a diagram showing a method for measuring the speed of a vehicle traveling straight in an oblique direction on the image pickup screen. 83 in the figure
Is a region of the vehicle F, 84 is a local region f for labeling, and 85 is a center of gravity WF (WFx, WFy) of the vehicle F. In the velocity measurement in this case, either the x coordinate or the y coordinate may be used.

Similarly, another embodiment will be described. Figure 1
2 is a diagram showing a first example in the case where the speed measurement section is not a straight line, where 86 is the region of the vehicle G, 87 is the local region g for labeling, and 88 is the center of gravity WG (WGx, WGy) of the vehicle G. Represent In this case, at the speed measurement start point s, the y coordinate is
The x coordinate is used at the speed measurement end point e. FIG. 13 is a diagram showing a flow of processing in the example of FIG. The flow of this processing will be described below according to the processing (a) to (h) shown in FIG.

(A) The frame number counter FN is set to 1. (b) Input the y coordinate WGy of the center of gravity of the area 86 of the vehicle G. (c) If WGy is greater than or equal to the y coordinate sy of the speed measurement start point, proceed to processing (d). In other cases, transition to the next processing frame image,
In the processing (b) state, the input of the y coordinate WGy of the center of gravity point of the next processing frame image is waited for.

(D) Transition to the next processing frame image. (e) Wait for the input of the x coordinate WGx of the center of gravity of the area 86 of the vehicle G. (f) The frame number counter FN is incremented by 1.

(G) WGx is the y coordinate ex of the speed measurement end point
If it exceeds, proceed to processing (h). Otherwise, proceed to process (d). (h) Input the value of the frame number counter FN into the distance frame number K.

The distance y (m) between s and e is not a straight line distance but a distance along the road from the speed measurement start point s to the speed measurement end point e. Similarly, another embodiment will be described. FIG. 14 is a diagram showing a second example of the method for measuring the speed when the speed measurement section is not a straight line.
9 is a region of the vehicle H, 90 is a local region h for labeling, 91
Represents the center of gravity WH (WHx, WHy) of the vehicle H. In this case, the x coordinate is used at the speed measurement start point s, and the y coordinate is used at the speed measurement end point e. FIG. 15 is a diagram showing a flow of processing in the example of FIG. The flow of this processing will be described below according to the processing (a) to (h) shown in FIG.

(A) The frame number counter FN is set to 1. (b) Input the x coordinate WHx of the center of gravity of the area 89 of the vehicle H. (c) If WHx is greater than or equal to the y coordinate sx of the speed measurement start point, proceed to processing (d). In other cases, transition to the next processing frame image,
In the processing (b) state, the input of the x coordinate WHx of the center of gravity point of the next processing frame image is waited for.

(D) Transition to the next processing frame image. (e) Wait for WHy input. (f) The frame number counter FN is incremented by 1.

(G) WHy is the y coordinate ey of the speed measurement end point
If it exceeds, proceed to processing (h). Otherwise, proceed to process (d). (h) Input the value of the frame number counter FN into the distance frame number K.

The number K of distance frames thus obtained is input to the speed calculation processing section 6. If the speed of the vehicle to be obtained is vkm / hour, the time for one frame is 1/30 second, so v is as follows.

V = 108y / K (2) The speed calculation processing section 6 calculates the above equation (2) and outputs the speed calculation result. Although the processed image is handled in frame units in the present embodiment, it goes without saying that it can be handled in field units in order to increase the time resolution of the speed measurement result. When image processing is performed in field units, the number of pixels processed is 768 × 240, and the number of lines in the vertical direction is halved, but the time resolution is doubled. At this time, the obtained speed v is as follows.

V = 216y / K (3) The target vehicle speed can be obtained by associating the speeds obtained by the equations (2) and (3) with the vehicle on the image.

Needless to say, in the speed measuring device requiring the time resolution higher than the above, the time resolution is increased when the high speed camera is used as the image input device. As is clear from this result, in the present invention, compared with the conventional technique, speed measurement can be performed without being aware in advance that the vehicle under measurement is in the speed measurement state, and a plurality of vehicles are connected in the measurement area. Even if it intrudes, the speed of each vehicle can be measured as long as the objects do not overlap, the measurement can be performed parallel to the traveling direction of the measurement vehicle, and the average speed in the non-linear measurement section can be measured.
There were improvements such as.

Although the embodiment of the present invention has been described with respect to the automatic speed measuring device for a vehicle, the present invention is not limited to this, and can of course be applied to a speed measuring device for other moving objects.

[0057]

As described above, according to the present invention, the vehicle is tracked on the image captured by the camera, and the speed is measured from the number of frames (or the number of fields) in which the object has passed a known range of distance in advance. Since the system is used, the speed can be measured without being aware in advance that the vehicle under measurement is in a speed measurement state, and even if multiple vehicles in a row enter the measurement area, the objects do not overlap. As long as the speed of each vehicle can be measured, the effect that it can be measured parallel to the traveling direction of the measurement vehicle, and the effect that the average speed in the non-linear measurement section can be measured can be obtained. as a result,
With the automatic speed measurement device, speed measurement can be performed without being aware in advance that the vehicle under measurement is in the speed measurement state, and speed measurement can be performed on multiple vehicles that continuously enter the speed measurement area. It has the advantages that it is possible, and that measurement in a direction perpendicular to the traveling direction or in a non-linear section is possible, which gives flexibility in setting the measurement location. Further, since the reflected wave of the microwave is not used, there is an advantage that the speed can be measured even if the outer skin of the object to be speed-measured is not metal.

[Brief description of drawings]

FIG. 1 is a block diagram illustrating an embodiment of the present invention.

FIG. 2 is a diagram showing a coordinate system of an image.

FIG. 3 is a diagram illustrating in detail a moving area extraction processing unit shown in FIG.

FIG. 4 is a two-lane three-lane road imaged from the front in the traveling direction.
It is a figure which shows the example of a value image.

5 is a diagram showing an image after several to several tens of frames from the image shown in FIG.

FIG. 6 is a diagram illustrating a method of tracking a moving object using a labeling process.

FIG. 7 is a diagram showing an example of a coordinate value input device using a pen-shaped light receiving element.

FIG. 8 is a diagram showing a flow of processing in a predetermined range passage determination processing unit.

FIG. 9 is a diagram showing a method of measuring the speed of a vehicle from a direction perpendicular to a road.

10 is a diagram showing a flow of processing in the example of FIG.

FIG. 11 is a diagram showing a method of measuring the speed of a vehicle traveling straight ahead in an oblique direction in an image pickup screen.

FIG. 12 is a diagram showing a first example when a speed measurement section is not a straight line.

13 is a diagram showing a flow of processing in the example of FIG.

FIG. 14 is a diagram showing a second example of a method for measuring the speed of a vehicle when the speed measurement section is not a straight line.

15 is a diagram showing a flow of processing in the example of FIG.

FIG. 16 is a schematic diagram of a conventional automatic velocity measurement method for emitting a microwave and measuring the velocity from the reflected wave.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Imaging unit 2 Moving region extraction processing unit 3 Moving region feature point calculation processing unit 4 Default range passage determination processing unit 5 Known point coordinate input processing unit 6 Speed calculation processing unit 20 Background image memory unit 21 Difference processing unit 22 Binarization processing Section 50 display display 51 pen-shaped light receiving element 52 coordinate value calculation processing section

Claims (3)

[Claims] 1. An automatic velocity measuring device for automatically measuring the velocity of a moving object moving in an image from an image signal input from an image input device, and means for extracting a region of the moving object and the extracted portion. Means for calculating the characteristic points of the region of the same moving object for each frame while tracking the moving object, and the coordinates of the speed measurement start point and the speed measurement end point where the distance of the section for measuring the speed is known on the image A means for inputting as a value and the number of frames or fields required for the calculated area feature point coordinates on the image of the moving object to pass between the speed measurement start point coordinates and the speed measurement end point coordinates An automatic speed measuring device comprising: means and means for calculating the speed of the moving object from the calculated number of frames or fields and the distance of a section for measuring the speed. 2. The automatic velocity measuring device according to claim 1, wherein the means for extracting the region of the moving body is a means for storing a background image signal in a memory in the absence of the moving body in advance, and the accumulated means. An automatic speed having a means for calculating a difference between a background signal and an input image signal in which a moving object exists, and a means for binarizing the calculated difference signal by a preset threshold value. measuring device. 3. The automatic velocity measuring device according to claim 1, wherein the means for calculating the characteristic points of the region of the moving body while tracking the moving body is a center of gravity of the region as the characteristic points of the region of the moving body. An automatic velocity measuring device, characterized in that it is arranged to track the animal body using points.