JPH0337355B2 - - Google Patents

Description

[Detailed Description of the Invention] (Technical Field) The present invention relates to an abnormality monitoring device that uses hue values as image data to have the same detection ability in bright areas and dark areas. It is particularly suitable for monitoring the occurrence of an abnormality in a monitoring area including a part.

(Background Art) Generally, current abnormality monitoring devices for intruder detection, crime prevention, and the like employ a method of detecting changes in brightness within a monitoring area using a black and white television camera. FIGS. 4a and 4b are diagrams for explaining the operation of the conventional example, and show input images captured using the conventional abnormality monitoring device. FIG. 4a is an example of an image in which there are people in front of a building, and sunlight is shining from the right front. Furthermore, FIG. 4b shows a person hiding behind a building. As can be seen by comparing Figure 4 a and b, there is a large difference in detection ability when the detection object is present in the bright area of the monitoring area (Figure 4 a) and when it is present in the dark area (Figure 4 b). The problem is that there are differences.

In particular, the image processing method in such a system is to find the brightness difference between the input image and the reference image for each pixel, count the number of pixels for which the brightness difference exceeds a predetermined threshold, and calculate this counted value. Generally, abnormalities are determined based on the size of the
In this method, if a detection object exists in a bright area, the contrast is high and the difference from the reference image is large, but in a dark area, the contrast is low and the difference from the reference image is small, making it difficult to detect the object. The image processing exemplified above is a type of binarization processing, but even when shading processing such as differentiation is used as another method, the differential value also depends on the contrast, so
There is a similar problem in that the detection ability in dark areas is inferior to that in bright areas.

(Object of the invention) The present invention has been made to solve the above-mentioned problems, and its purpose is to:
An object of the present invention is to provide an abnormality monitoring device in which changes in detection ability due to differences in local illumination light amounts in a monitoring area are reduced.

(Disclosure of the Invention) As shown in FIGS. 1 and 2, the abnormality monitoring device according to the present invention includes an image input means 1 for capturing an image of a monitoring area and quantizing an image signal; An image processing means 2 that compares the image obtained by the above with a reference image to obtain information necessary for abnormality determination, an abnormality determination means 3 that determines an abnormality based on the information obtained by the image processing means 2, and an abnormality determination means 3 that determines an abnormality based on the information obtained by the image processing means 2. In the abnormality monitoring device, the image input means 1 includes a color image capturing means 11 such as a color television camera, and a color image signal obtained by the color image capturing means 11. and a hue extraction means 12 for inputting the image to the image processing means 2.

In the present invention, only the hue is extracted as image data, so the detection is almost the same whether the detected object is in a bright area or a dark area, as illustrated in Figure 3 a and b. Therefore, the detection ability is less likely to change due to differences in the amount of illumination light in the monitoring area.

Preferred embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a block diagram showing a schematic configuration of an abnormality monitoring device according to an embodiment of the present invention.
Color image capturing means 1 consisting of a color television camera
1 three color signal outputs (R: Red, G: Green,
B: Blue) is input to the hue extraction means 12, and the hue is determined. Here, hue refers to G/R,
It is a numerical value that represents the color itself, which does not depend on the amount of light, and can be expressed by a formula such as R/(R+G+B) or G/(R+G+B). The output of this hue extracting means 12 is input to the main body 5 of the abnormality monitoring device, and conventional processing as described below is performed.

FIG. 2 is a block diagram showing an example of the internal configuration of the main body 5 of the abnormality monitoring device. The image processing means 2 has an input image memory 21 and a reference image memory 22. The reference image memory 22 stores image data in which images of the monitoring area when there is no abnormality are taken in as reference images at regular update intervals. The input image memory 21 stores hue images of the current monitoring area that are taken in from the image input means 1 from time to time. The absolute difference circuit 23 includes both image memories 21,
22 luminance data is subtracted between pixels to convert it into an absolute value. The binarization circuit 24 binarizes the output of the absolute difference circuit 23 using an appropriate threshold value. As a result, an image consisting only of pixels whose brightness has changed by a predetermined threshold value or more between the current image and the reference image is obtained. The abnormality determination means 3 receives the binarized signal obtained by the binarization circuit 24 and detects an abnormality from this signal. For abnormality detection, for example, a method can be used in which the signals obtained from the binarization circuit 24 are counted, and if the count exceeds a certain value, it is determined that an abnormality has occurred. Also,
As a method for more advanced abnormality determination, a method having an inference function based on artificial intelligence can also be used. The output means 4 outputs a notification when an abnormality is detected, and can be of various types such as a monitor television, a printer, an alarm lamp, an alarm buzzer, etc.

In the present invention, a hue signal is used as the image signal input to the main body 5 of the abnormality monitoring device, instead of the image signal from a black and white television camera. The calculation in the hue extraction means 12 may be performed in analog or digital form, but ultimately it is input to the input image memory 21 of the image processing means 2 as a digital signal. Examples of image input of the monitoring area using this hue signal are shown in FIGS. 3a and 3b. The scenes shown in Figures 3a and b are the same as those shown in Figures 4a and 4b.
This corresponds to case b. Since it is a hue-only image, there is no effect of the shadow of the building, and parts of the building that have the same hue have the same value. Regarding the detected object,
The contrast remains constant even under conditions where uneven illumination exists. In this way, it is possible to make the detection capability of the abnormality monitoring device substantially uniform over the entire monitoring area. Strictly speaking, the hue under direct sunlight and the hue in the shade where there is no direct sunlight and only receives scattered light from the surroundings are slightly different, but there is still a difference in brightness of more than an order of magnitude under the same conditions. The detection ability can be sufficiently stabilized compared to those using general black and white images.

(Effects of the Invention) In the present invention, as described above, since abnormality monitoring is performed using image data consisting only of hues, for example, areas of a building that are exposed to direct sunlight and areas that are in the shade are monitored. This has the effect that stable detection ability can be obtained over the entire screen even when there is unevenness in the amount of illumination within the monitoring area.

Note that, as described in the description of the embodiments, the present invention is particularly suitable when using a configuration in which the reference image is updated in the image processing means.
That is, according to the present invention, since the input image is stable even if the amount of illumination light changes, the frequency of updating the reference image can be reduced.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a schematic configuration of an abnormality monitoring device according to an embodiment of the present invention, FIG. 2 is a block diagram showing a schematic configuration of an abnormality monitoring device main body in the same embodiment, and FIGS. 3a and b 4A and 4B are explanatory diagrams for explaining the operating state of the above embodiment, and FIGS. 4A and 4B are explanatory diagrams for explaining the operating state of the conventional example. 1 is an image input means, 2 is an image processing means, 3 is an abnormality determination means, 4 is an output means, 11 is a color image capturing means, and 12 is a hue extraction means.

Claims (1)

[Claims] 1. An image input means that images the monitoring area and quantizes the image signal, an image processing means that compares the image obtained by the image input means with a reference image and obtains the information necessary for abnormality determination, and a In the abnormality monitoring apparatus, the image input means includes a color image capturing means such as a color television camera, and a color image capturing means, such as a color television camera. An abnormality monitoring device comprising: hue extraction means for extracting only the hue from the color image signal obtained by the color image signal and inputting the extracted hue to the image processing means.