JP3217151B2 - Moving object detection method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving object detecting method, and more particularly to a moving object detecting method of appropriately moving a place such as various kinds of automatic transport vehicles.

[0002]

2. Description of the Related Art Conventionally, when inputting a machine vision image with a movement, particularly when a continuous image is required, the movement of the machine is temporarily stopped and the target image is continuously sent to a processing device. It takes in and performs necessary processing.

[0003] However, the above-mentioned conventional techniques have the following difficulties. That is, if the movement is stopped and the required processing is performed each time continuous image capture is required, the processing speed may be reduced. Also, stop seals the continuous image capture is not performed, such as in a state not, uniformly reliable and uniform treatment throughout the entire movement path is not performed.

[0004]

BRIEF attempt to resolve the object of this invention has been made to solve the above ISSUES, it is not necessary to stop the movement whenever in need of continuous image capture, the required processing is prevented from becoming low processing speed for applying the, evenly reliably and which, as a uniform processing possible, to provide a moving object detection method with excellent characteristics throughout the entire movement path With the goal.

[0005]

Moving object detection method according to [SUMMARY for a] The present invention has been made in order to fulfill the above object, a camera 11 for sequentially tracing the object body shall be the subject, the first
At the time point (e.g., t = 0) and the tracking mode (e.g., PAN, TILT, ZOOM)
) At a second time after a predetermined time (eg, t =
a synthesized image memory for predicting and storing the image data of t) ;
At the second time point, a comparison image memory for storing image data based on an image signal from the camera, an image stored in the composite image memory 14, and an image stored in the comparison image memory 15 , A processed image memory 17 for storing the result of the comparison by the comparing unit 16, and a result of the comparison stored in the processed image memory 17, and information corresponding to the recognition is recognized. and it is characterized in that it comprises a recognition unit 1 8 for transmitting to the control unit 13.

[0006]

In the moving object detection method according to the present invention, the moving object detection method is obtained from the camera 11 at a certain first time point (eg, t = 0).
Image and tracking information at the same time, e.g.
Mutual movement between the cameras (eg, PAN, TIL
T, ZOOM, etc.) based on the information from the first time
Prediction of an image appearing at a second time point (t = t) after a lapse of time
Then, the image is stored in the composite image memory 14. Also,
Compare the images actually obtained from the camera 11 at the second time point
Is stored in the image memory for use 15. And the comparison unit 16
The image stored in the composite image memory 14 and the image for comparison
The image is compared with the image stored in the image memory 15 and processed.
It is stored in the image memory 17. The recognizing unit 18 processes the processed image
The result of the comparison stored in the memory 17 is recognized. Therefore, according to the present invention, it is not necessary to stop the movement each time continuous image capture is required, and it is possible to prevent a processing speed for performing a required process from being lowered , and > It is possible to provide a moving object detection method with excellent characteristics that enables uniform and reliable processing over a physical movement path.

[0007]

FIG. 1 is a schematic block diagram of a moving object detection system according to an embodiment of the present invention. In FIG 1, the camera 11 is a type of image input devices, such as those moving to that object body interest (in practice the camera is moving) will sequentially tracking the aspects of . The A / D unit 12 is for converting an analog input image signal captured by the camera 11 at the preceding stage into a corresponding digital signal. Control unit 13, as well as communication with the drive unit 19 which is a part of the moving body is for commanding the operation of each species means. The composite image memory 14 stores the camera 11
From the image signal at the first time point (e.g., t = 0) and the mutual
Supports moving modes such as PAN, TILT, ZOOM, etc.
The second time point after a predetermined time based on a signal (t = t)
In order to predict the image data at and store the image data. Comparison image memory 15, the first
At time 2, image data based on the image signal from the camera 11 is stored. The comparison unit 16
This is for comparing the image stored in the composite image memory 14 with the image stored in the comparison image memory 15. The processing image memory 17 stores the comparison unit 1
6 to store the result of the comparison. The recognition unit 18 recognizes the result of the comparison stored in the processed image memory 17 and transmits information corresponding to the recognition to the control unit 13. The driving unit 19 is included with e.g. unattended <br/> mobile, such as robotic mobile device.

Next, the mode of operation of the system of the above embodiment will be described with reference to various examples of image data shown in FIGS. The image data input from the camera 11 at the present first time point t = 0 is digitally converted by the A / D converter 12 at the next stage, and then, based on the mutual movement mode between the camera and the object at that time point. An image at the expected second time t = t is generated and stored in the composite image memory 14. In FIG. 2, 1 is an effective image area, 2 is an image processing area, and 4A (car) and 4B (house) are target images. Information describing the mode of tracking
As a report, the control unit 13 determines the direction of movement from the drive unit 19 of an appropriate automatic transport vehicle as a moving body, by using PAN (horizontal movement), TILT (vertical movement), or ZOOM (simultaneous in both horizontal and vertical directions). (Enlargement or reduction), and operate various means corresponding to the information. The prediction and generation of the image at the second time point
For example, when the PAN is view based on the original information of FIG 3
(In the figure the horizontal movement to the right) of PAN synthesis operation of images is made. Further, when the TILT the synthesis operation of the images is performed (the vertical downward movement in Figure) TILT groups <br/> Hazuki Figure 4 the original information of FIG. In the case of ZOOM, FIG.
(In FIG simultaneous expansion in both the horizontal and vertical) ZOOM of basis Figure 5 to the original information of the synthesis operation of the images it is made. Image data from the second time point (t = t) of the camera 11, Ru stored in the specific較用memory 15. For example, when information PAN is received from the control unit 13,
As shown in FIG. 6, the composite image memory 14 has t
= Image when 0 is stored the image data in a state of moving horizontally, also images data from the camera 11 when of the comparison image memory 15 t = t is stored as it is. Then, the two image data are compared by the comparing unit 16 (for example, a comparison is made by subtraction as shown in FIG. 6), and the result is obtained as a difference value. Here, since the two image data are completely the same, for example, nothing is displayed on the processed image memory 17. Also,
As shown in Figure 7, 4 to the image data of the both
There is a different target image (abnormal target) C (person)
In this case, the corresponding image is displayed in the processed image memory 17. That is, since an image different from the expected image occurs, for example, the occurrence of an abnormal situation is recognized by the recognition unit 18, and a signal corresponding to the occurrence of the abnormal situation is output to the control unit 13. Become.

[0009]

As described above, the moving object detection method according to the present invention can be used at a certain first time point (eg, t = 0).
Image obtained from camera 11 and state of tracking at the same point
(Eg, PAN, TILT, ZOOM, etc.)
At a second time point (t =
Predict the image appearing at t) and store the image in a composite image memory
14 and the camera 11 is actually
The obtained image is stored in the comparison image memory 15. So
Then, the comparison unit 16 stores the data in the composite image memory 14.
Image and the image stored in the comparison image memory 15.
The image is compared with the image, stored in the processed image memory 17,
Reference numeral 18 denotes a comparison result stored in the processed image memory 17.
Recognize the fruit. Since the present invention is configured with such features, it is not necessary to stop the movement each time continuous image capture is required, and the processing speed for performing the required processing is prevented from being reduced. Thus, there is an effect of providing a moving object detection method having excellent characteristics such that uniform and uniform processing can be performed over the entire moving path.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of a moving object detection system according to an embodiment of the present invention.

FIG. 2 is an exemplary view of an image display for explaining an operation mode of the method of the embodiment.

FIG. 3 is an exemplary view of an image display for explaining an operation mode of the method of the embodiment.

FIG. 4 is an exemplary view of an image display for explaining an operation mode of the method of the embodiment.

FIG. 5 is an exemplary view of an image display for explaining an operation mode of the method of the embodiment.

FIG. 6 is an exemplary diagram of an image display for explaining an operation mode of the method according to the embodiment.

FIG. 7 is an exemplary view of an image display for explaining an operation mode of the method of the embodiment.

[Explanation of symbols]

11 camera; 12 A / D unit; 13 control unit;
Synthetic image memory; 15 comparison image memory; 16 comparison unit; 17 processing image memory; 18 recognition unit; 19 driving unit.

Claims (1)

(57) [Claims] 1. A camera for successively tracking an object body shall be the target, the image signal and the from the camera at the first time point add
Based on the aspect of the trace, the image data at the second point in time after the predetermined time has elapsed.
A composite image memory for predicting and storing data, a comparison image memory for storing image data based on an image signal from the camera at the second point in time, an image stored in the composite image memory, comparing unit for comparing the images stored in the use image memory, processing an image memory for storing the result of the comparison by the comparison unit, and a recognition unit, which recognizes the result of the comparison stored in the processing image memory Moving object detection method comprising.