JPH03293884A - Mobile body detector - Google Patents

Abstract

PURPOSE:To prevent erroneous discrimination performed by fetching the whole mobile body in image memory storing a change area at a time when the partial change area of the mobile body appears on the edge of a display screen. CONSTITUTION:An inputted image is stored once in the image memory 2, and is displayed on an image display part 3, and also, a differential binary image extraction part 4 extracts a differential binary images from all the image areas, and stores them in the image memory 5, and a mobile body detecting and judging part 6 detects and judges whether or not an area equivalent to the mobile body exists. The size of PM2.5 of the image memory for processing storing an input images and the differential binary images is set sufficiently larger than that of DM of the image memory for display, and the device is comprised so that a real partial binary image can fetch the whole shape of an original mobile body when the change area by the mobile body appears at either the four sides of the edges in the neighborhood of the screen. Thereby, erroneous judgement can be prevented.

Description

Detailed Description of the Invention [Objective of the Invention] (Industrial Application Field) The present invention is an image monitoring device that detects illegal intruders using images taken by a television camera, for example, and detects moving objects from input images. The present invention relates to a moving object detection device that performs detection.

(Prior Art) In recent years, for example, in security systems, image monitoring devices that detect illegal intruders by using images captured by a television camera have become widespread. This is partly because low-cost devices have become possible due to recent remarkable developments in semiconductor technology and the spread of digital image processing technology.

In such an image monitoring device, for example, in order to detect an illegal intruder, it is necessary to extract information on changes caused by the movement of the illegal intruder within the image. Conventionally, as a method for extracting such a changed area, a moved portion (changed area) was extracted by simply performing a subtraction process with a temporally distant image on the display screen.

(Problem to be solved by the invention) However, in the conventional method of extracting change regions,
An object that enters the image of which the imaging area is being monitored will always be partially caught at the peripheral edge of the screen at first. When the above moving object is detected at the peripheral edge of the screen,
The area of change will be partial. On the other hand, when determining what the change area is, there is a risk that the determination may not be complete or that erroneous determination may occur. Furthermore, this makes the process for this determination complicated.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a moving object detection device that can perform processing similar to change area detection processing within the screen even for a partial change area of a moving object that appears at the edge of a monitoring screen. With the goal.

[Structure of the Invention] (Means for Solving the Problems) A moving object detection device of the present invention includes: an image input means for inputting an image; a display means for displaying the image input by the image input means; a changing area extracting means for extracting an image indicating a changing area of a moving object by accepting an image inputted by the input means and performing arithmetic processing with a temporally distant image; an image memory for storing a captured image; and a moving object detection means for detecting a moving object from the image stored in the image memory, and the size of the image memory is compared with the image display screen size of the display means. It is characterized by its large size as it has a certain area around the display screen.

(Function) In the moving object detection device according to the present invention, when a changed area of a part of the moving object appears at the edge of the display screen, the entire moving object is already captured in the image memory that stores the changed area. , in response to this, a changing area is extracted and a moving object is detected.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 shows a moving object detection device in an image monitoring device for detecting illegal intruders, etc. according to the present invention. That is, the image input unit 1 inputs an image captured by a television camera, for example, of a monitoring area. This input image is temporarily stored in the image memory 2 and then displayed on the image display section 3. The differential binarized image extraction unit 4 performs subtraction processing between three temporally consecutive images stored in the image memory 2, binarizes them at a specified threshold, and generates a differential binarized image over the entire image area. The images are extracted and stored in the image memory 5. The moving object detection/judgment unit 6 uses the image memory 5
Detects and determines whether an area corresponding to a moving object of a preset size exists in the differential binarized image stored in the , and if it detects and determines that it exists, indicates the corresponding area. The image is displayed on the image display unit 3 in the form of Here, as a method of indicating the corresponding area, for example, the area determined to be a moving object may be displayed in a colored manner, or the area determined to be a moving object may be displayed as a high-intensity rectangle.

Here, in the case where the entire moving object is completely within the screen, a generally used hardware configuration is sufficient for the above-mentioned moving object detection processing, but in this embodiment, the processing shown in FIG. As shown in a), a processing image memory (image memory 2.
5) The relationship between the size of PM and the size of the display image memory DM (provided in the image display section 3) is such that the size of the processing image memory PM is sufficiently larger. That is, the size of the processing image memory PM is larger than the size of the display image memory DM by the amount of a certain area around the display image memory DM. In other words, the size of the processing image memory PM is larger than the image display screen size of the image display section 3 by the amount of a certain area around the display screen.

As a result, when the area of change caused by the moving object touches any of the four edges around the screen, the actual difference binarized image captures the entire original shape of the moving object. Therefore, compared to the conventional case where the size of the processing image memory and the display image memory are the same, there is no need to worry about misjudgment when detecting a moving object. For example, FIG. 2(b) shows a changing area acquired at a certain arbitrary time when a person is intruding from the right edge of the screen.

At this time, only the left half of the person's original shape is extracted on the screen. If region integration processing is performed on this changed region, there is a risk that it will be extracted even though it is a person, and as a result of the region integration processing, it will be considered as a noise component because the width size is small.

On the other hand, if the processing image memory is large enough to include the display image memory, changes in the overall size of the original person, including the change area indicated by the dotted line, will be recorded on the processing image memory. Since the region can be extracted, it can be determined that it is a human region by integrating the changed regions.

In addition, as shown in Figure 2 (C), when a large moving object enters from the right edge of the screen, the change extracted and displayed at the right edge of the screen is compared to the extraction result of the change area at a certain arbitrary time. As a result of integrating a portion of the area, there is a risk that it may be mistakenly determined to be a human area. On the other hand, if the processing image memory is large enough to include the display image memory as described above, the entire original track including the changed area indicated by the dotted line will be stored on the processing image memory. Since a region of change in size can be extracted, it is possible to determine a moving object larger than a person by integrating the region of change.

Note that the present invention is not limited to the above embodiments.

For example, with the configuration of the embodiment described above, it is possible to simplify and speed up processing by checking the amount of change only in the peripheral edges of the display screen, and if a change is detected, starting a moving object detection algorithm. can.

Furthermore, it is also possible to not perform the detection process if there is a change in the peripheral edge of the display screen.

Furthermore, similarly, if there is a change in the peripheral edge of the display screen that is greater than a threshold value, it may be possible to display the changed side in color to alert the observer.

[Effects of the Invention] As detailed above, according to the moving object detection device of the present invention, the detection process can be started when the moving object to be detected partially appears on the edge of the display screen, and the Even if the changing area of a part of the moving object appears on the screen, the entire image of the moving object is captured in the image memory that stores the changing area, so there is no need for special complicated processing. ° Processing similar to normal movement and object detection processing within the screen can be performed.

[Brief explanation of drawings]

The figures are for explaining one embodiment of the present invention. Figure 1 is a block diagram showing the overall configuration, and Figure 2 (a) shows the relationship between the sizes of the processing image memory and the display image memory. The figure shown in Figure 2(b) is a diagram for explaining the processing when a part of a person is extracted as a moving object at the edge of the screen.
Figure (C) is a diagram for explaining processing when a part of a large truck is extracted as a moving object at the edge of the screen. 1... Image input section, 2, 5... Image memory, 3...
・Image display section, 4...Difference binarized image extraction section, 6...
- Moving object detection/judgment unit, PM... image memory for processing, DM... image memory for display.

Claims (1)

[Claims] Image input means for inputting an image; display means for displaying the image input by the image input means; a changing area extracting means for extracting an image showing a changing area of a moving object by performing arithmetic processing; an image memory for storing an image obtained by the changing area extracting means; moving object detection means for detecting a moving object from an image, and the size of the image memory is larger than the image display screen size of the display means by a certain area around the display screen. A moving object detection device characterized by: