JPS62147888A - Picture monitoring system - Google Patents

Abstract

PURPOSE:To exactly detect a variation in illuminance due to an abnormality such as a trespasser or fire by not regarding a drastic variation in the external light such as illumination and sunshine incident to a monitoring area as an abnormal variation. CONSTITUTION:Three kinds of systems to detect a trend of illuminance-variation are provided, and switching between them is executed by an illuminance variation detection system setting circuit 6. The first system is such that the output variation in an illuminance meter 7 is obtained by an illuminance variation calculating circuit 8 for the switching. The second system is such one that a histogram of the magnitude of the variation in each picture element is generated by a histogram generating circuit 9. And the third one is one that uses a consecutive-variation decision circuit 10 which decides the consecutive variation in plural picture elements. Of the first system, a variation amount (deviation) (or the proportion of the illumination of the preceding picture to that of a reference picture) between the illuminance of the current picture detected the meter 7 at the time when fetching the picture and that of the reference picture (or the preceding picture) at the time of fetching it, is calculated by an illuminance-variation calculating circuit 8. In case the resulting illuminance variation amount (deviation) is larger that a set value, it is not regarded as an abnormal variation, but instead, the reference picture is updated with the current picture by a reference picture updating circuit 24.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to an image monitoring system that issues an alarm by detecting a change in a monitoring image by a TV camera.

[Backweight technology] The present inventors have already proposed a method in which changes are detected by projecting a difference image between the current image and a reference image, and the image information of the changed area is compressed and transmitted in a narrow band. The image projection increases with changes in illuminance in the monitoring area, so even if there is no abnormality such as an intruding object or fire, it is possible to detect changes in the image due to lighting on/off or changes in sunlight and transmit the changed image. There was a so-called false alarm, and it was necessary to check the transmitted image each time to confirm whether it was a false alarm or not.

Therefore, the present inventors have proposed a method in which the contours and edges of an image are obtained as a line drawing by differentiation and binarization of the image, and changes are detected based on the differences in the line drawings to remove the effects of turning on and off the illumination.゛Currently, line drawing creation,
Computer visitor (population intelligence frB) processing requires a large amount of calculation, so it requires a large capacity storage device (intelligence base) and multiple image processing processors, making the system expensive. At present, there are still problems with practical application.

[Objective of the Invention 1] The present invention has been made in view of the above-mentioned problems, and is intended to prevent sudden changes in outside light incident on a monitoring area such as lighting or sunlight from being regarded as abnormal changes. An object of the present invention is to provide an image monitoring method that can reliably detect changes caused by an intruder or an abnormality such as a fire.

Margin below [Disclosure of invention] The present invention will be explained below with reference to Examples.

FIG. 1 shows the overall configuration of this embodiment, and shows a TV camera 1 using an image sensor such as a CCD that is sensitive to infrared regions.
The TV camera 1 is attached to the ceiling of the monitoring area, as shown in FIG. 4, to take images of the monitoring area. The TV camera 1 can select whether the image to be captured is in the infrared region or the visible region by switching the bandpass filter 32. Since the A/D converter 2 is used to digitize the video signal from the TV camera 1, the digitized image data is stored in the 7 frame cross 77. This memory is called incorporation,
Normally, the time required to capture this image data is set to about 0.5 seconds for normal intruder detection, although it depends on the moving speed of the object to be detected.

In the embodiment, there are three 7-frame crosses 7a, the current image 7-frame cross 7a3 is for storing the latest image data, and the reference image frame buffer 75 is for storing the background image of the monitoring area where there are no intruders etc. It is for storing data. The previous image 7 frame x 7 a 4 is for storing the previous captured image.

By the way, there are two types of image change detection: comparison between the current image and a background reference image, and comparison between the current image and the previous image.In the former, even if the intruder is stationary, it is considered that there is a change. The entire intruder is detected as a changed area, but in the latter case, when the intruder moves, only the moved part is considered a changed area, and the intruder's position in the previous image and the current image are different. If they are far apart, the position of the intruder in the previous image becomes a disappearing change area in the current image, and the position of the intruder in the current image becomes an appearance change area. Therefore, when comparing the background reference image and the current image, only the appearance change area is included.

The embodiment of the present invention mainly performs image comparison normalized to illuminance changes in order to detect abnormal changes by comparing the current image with a background reference image, but when it seems that the illuminance has changed, image comparison is performed again. By doing this, changes in illuminance are not considered abnormal changes,
This is related to an image monitoring method that updates the background reference image, but it also detects signs of illuminance changes when comparing the previous image and the current image, performs image comparison again, and normalizes the illuminance. Of course, it is also possible to do the same thing.

In this embodiment, three types of methods for detecting signs of a change in illuminance are employed, and the method can be switched by an illuminance change detection method setting circuit 6.

First, to give an overview of the three types of methods, f51
The second method is a method in which a change in the output of the illuminance meter 7 is determined by an illuminance change calculation circuit 8, and the second method is a method in which a histogram of the magnitude of change for each pixel is created in a histogram creation circuit 9. The method (4) uses a continuous change determination circuit 10 that determines continuous changes in a plurality of pixels.

The method using the illumination meter 7 detects the amount of change (deviation) between the illuminance at the time of capturing the current image (Xi) detected by the illumination meter 7 and the illuminance at the time of capturing the reference image (or previous image) (or the amount of change (deviation) between the illuminance at the time of capturing the reference image (or previous image) illuminance change calculation circuit 8
Calculate with. At this time, if the amount of change in illuminance (deviation) is larger than a certain set value, the change is not considered abnormal, and the reference image updating circuit 24 updates the reference image with the current image. Also, from the calculated value of the illuminance change calculation circuit 8, a calculation circuit that calculates a correction coefficient (function) for correcting each pixel of the reference image so that it has the same conditions as the illuminance of the current image, and calculates the magnitude of change for each pixel. Based on the calculation of step 26, the correction coefficient (function) determined by yFc is registered in the correction coefficient (function) buffer 12 for each pixel by the correction coefficient (function) calculator @11.

Then, the illuminance normalization calculation circuit 13 multiplies the product of the registered correction coefficient and the calculated change amount for each pixel of the reference image, or multiplies a function for each pixel of the reference image to obtain the current image. The illuminance is normalized in comparison with the reference image. Of course, the correction amount may be decreased or added.

Here, when the auto iris circuit or automatic gain adjustment circuit of TV camera 1 is working, the correction coefficient is such that the average brightness of the entire image is kept almost constant, so the portion that becomes a shadow when the illuminance is high is a negative correction coefficient. In other words, after making the average value the same, the brightness of each pixel of the image is corrected. On the other hand, in the case of an image that is not affected by the auto iris circuit or the automatic gain $I adjustment circuit, the correction coefficient is
Corresponds to the reflectance of light in one direction of the camera, for example, in the second direction.
After the correction calculation is performed by the correction calculation unit 13a of the illuminance normalization calculation circuit 13 configured as shown in the figure, the DIN Wj4 adjustment unit 1
If adjustment is performed in 3b, the auto iris circuit and automatic gain adjustment circuit can be simulated.

In addition, if the lighting conditions of the monitoring area change, the shadow area will also change, so a correction coefficient (function) must be calculated each time, but if the lighting conditions are constant, it is necessary to This will prevent false alarms caused by the influence.

Next, detection of signs of illuminance changes using a histogram will be described. First, the histogram of changes for each image
Create a histogram using the histogram (frequency distribution) of the magnitude of change for each pixel of the difference image obtained by calculating the Sabahigi value of the 71 image for each pixel of the change detection grid. A circuit 9 creates a histogram of the magnitude of change for each pixel. If there is a change in illuminance, it will change throughout the image, so the proportion (or number) of pixels (zero pixels) with a change smaller than a certain setting value will be small, so if we test this condition, This allows the change detection process by capturing the next image to be skipped. This change detection skip is performed by the reference image update determination circuit 25. Of course, the method using a histogram is included in the verification of the magnitude of change for each IW element to obtain the projection described later, and when the proportion of zero pixels is small, it is considered a change in illuminance and is transferred to the next change detection process. can.

Now, in the example of FIG. 1, if normalization of illuminance changes is not performed, the current image is the illumination off image as shown in the photograph in FIG. 14, and the illumination on image as shown in the photograph in FIG. In an example where the image is used as a reference image, if you compare the histogram of the current image (shown in Figure 16) and the histogram of the reference (back image (shown in Figure 18)), you will see that there is a large difference in the average value. The average value of the histogram of the current image is 42, and the average value of the histogram of the reference image is 142.The histogram of the difference image in this case is as shown in FIG. Setting value R1 for testing the magnitude of change for each pixel in the difference image (with the threshold value in Figure 20, the number of pixels (zero pixels) with a magnitude of change smaller than j (2) is smaller than a certain value R6 Sometimes, it determines that there is a sign of a change in illuminance, and gives a command to the reference image update circuit 24 to update the reference image to the current image. The histogram looks like the fJS33 diagram, where most of the pixels are zero pixels below the set value R1, and the number of zero pixels exceeds the set value R6.

Furthermore, in the case of intrusion detection examples such as the photographs in Figures 5 and 6, the histogram (Figure 7) of the current image (Figure 5) and the histogram (Figure 9) of the reference image (Figure 6) are There is almost no difference in shape or average value. This 1 Sat TV camera 1
The same is true when the auto iris (automatic gain adjustment circuit) of

In the histogram of the difference image in this case (FIG. 11), the number of zero pixels below the set value R1 is greater than the set value R6, so it can be determined that there is no sign of illuminance change.

Note that when a monitoring area is set by the illuminance change monitoring area setting circuit 27 and the number of change detection grids in this area changes, it is necessary to normalize the frequency of the histogram and the set value R6 to the number of change detection grids. There is.

In this way, by detecting changes in illuminance using this histogram, false alarms due to changes in illuminance can be prevented without providing the illuminance meter 7.

Next, a method for determining signs of changes in illuminance without using the illuminance meter 7 or histogram will be described. In this method, if the continuous change determination circuit 10 determines that the number of continuous changes is greater than or equal to the set value R7, it is determined as an abnormal change and an alarm is issued; if the number of changes is less than the set value R7,
When it is determined that there is a sign of a change in illumination, a command to update the reference image is given to the reference image update circuit 24, and the reference image is updated to the current image. This method of detecting signs of illuminance changes is effective when the entire image changes in a short period of time, such as when a light is turned on and off, but when the change is relatively gradual, such as a change in solar radiation, it is possible to The difference between the current image and the reference image may increase over a considerable period of time and may be judged as an abnormal change at some point. It is also necessary to calculate a moving average and use the moving average as a reference image, or to update the reference image at regular intervals.

Each of the above methods for detecting signs of illuminance changes is effective even when the monitoring area is indoors and only a portion of the image is illuminated by sunlight entering through a window.In particular, the histogram-based method detects changes in illuminance. By setting the pixel area used for this purpose as an illuminance change monitoring area only in the position illuminated by sunlight using the illuminance change monitoring area setting circuit 27, it is possible to determine whether there is an illuminance change in the histogram of the difference image of only this area. If the determination is made, the reference image may be normalized.

In addition, when there is a low possibility that only a part of the image is illuminated and the lighting conditions are fixed indoors, or when used outdoors, the reference image or current image is normalized against changes in illuminance across the entire monitoring area. By doing so, it is possible to more accurately detect changes even in the case of gradual changes in illuminance.

Next, the change detection process of this embodiment will be explained. First, the pixel change test circuit 28 tests the magnitude of change for each pixel using a set value R1, and the projection calculation circuit 14 calculates the magnitude of only the pixels with a change larger than the set value R1 for each line parallel to the X and Y coordinate axes. The projection is calculated by adding up each time. Figure 12 shows a binary image of the changed pixels, which is obtained by comparing the current image in Figure 5 with the reference image in fjS6 and binarizing the changed pixels into 1 and the zero pixels into 0. Detection grid (every 4 pixels)
The change pixels are shown as points, and the outline of the intruder appears. Incidentally, FIGS. 8 and 10 show buffer images of the binary images shown in FIGS. 5 and 6. For binary images, the contour image is shaped by propagation and degeneration processing in known processing directions, and is used as an image feature for pattern recognition and alignment. It is possible to do this by finding the following information, and it is used in conjunction with the identification of intruding objects. Therefore, the illumination conditions important for the identification of intruding objects are kept substantially constant by normalization to illuminance,
The reference value R1 for changed pixels and zero pixels is clearly defined, making it easier to extract binary pixels from changed pixels, ensuring the identification of intruding objects, fire, etc., and enabling image monitoring with fewer false alarms of abnormal change determination. It has the effect of becoming Further, since the correction coefficient for the background is applied to the intruding object, mismatching of the correction coefficients makes the pixel change of the intruding object noticeable, so that it is easy to binarize the changed pixels. However, in order to verify an intruding object, correction is performed to match the intruding object. If there is a lot of noise and it is difficult to binarize changing pixels, setting an upper limit on the size of changing pixels will allow the size of changing pixels to be moderate and at the same time reduce large differences due to isolated point noise. The influence of pixel value changes can be reduced.

Figure 13(a) shows
The projection on the axis is shown, and FIG. 13(b) shows the projection on the Y axis. In the figure, the one-dot chain line indicates the set value R2, and the two-dot chain line indicates the set value R5. The projection of this changing pixel size is 'RR
Since it is based on the difference between the object and the intruding object, it is not suitable for identifying an intruding object, but projection of a binary image of changed pixels is useful for identifying a changed object when there is only one changing object. When there are a plurality of changing objects, the changing object is separated and cut out at the level of the binary image of changing pixels as shown in FIG. Verification in the object verification identification circuit 16
W& will be helpful. Here, the changing object matching and identifying circuit 16 has a standard pattern 22 stored in advance, and can perform comparing and identifying the changing object by comparing this standard pattern 22 with cutting data of the changing object and projection calculation data.

In history, the values of changing pixels in the difference image are replaced with the corresponding pixel values of the current image to calculate the projection, and the changing object is compared and
Identification can be made in more detail.

In other words, in the case of a fire, the value of the changed pixel increases (ignitability), so the strength of the projection also increases relative to the size of the changed area on the coordinate axes, and conversely, the strength of the projection of an intruder, etc. is small. Become. In other words, when calculating the projection on each coordinate axis, the value of the projection of the changing pixel of the current image is divided by the cumulatively added number of changing pixels (same as the projection of the binarized changing pixel), and the brightness of the changing object per changing pixel is calculated. If the brightness of the object changing per one changing pixel is abnormally high, it can be said that there is a high possibility of a fire. The brightness/darkness determination circuit 17 performs normalization and brightness/darkness determination for each pixel that changes in projection.

The above explanation is for a 7/0 image, but in the case of a color image, the images are compared for each color component of RGB, the changed pixels are binarized, and after the binarization, the binarized changed pixels are The changed pixels are determined by superimposing the pixels.

In addition, in this embodiment, in order to strictly distinguish between a man-made disaster and an intruder, the bandpass filter 32 separates the infrared region and the visible light region, and simultaneously and separately A/D converts the light into the visible light region in the case of an intruder. The two are distinguished by the fact that the change is mainly in the light range, and in the case of flammable objects, the change in the infrared range is large, but incandescent light bulbs and sunlight also have a large change in the infrared range. By similarly normalizing the change in illuminance in the outside area, false alarms can be prevented.

Now, testing the change area by projection has the advantage that it is less affected by noise such as isolated points of change pixels and can clearly detect changes than determining the change range using a binary image of change pixels as shown in the fj412 diagram. There is. Furthermore, when there are a plurality of changing objects, the change detection surface area can be separated more accurately by cutting out the changing objects using a binary image of changing pixels and then calculating the projection.

Here, in the changing area projection verification circuit 18, the set value [<2
The change detection area verification circuit 19 calculates the range on the coordinate axes of the projection beyond , for each coordinate axis (X, .

X, 2)-(Xn1-Xn2)-(Y, 1.Y,
2) - (Yn,, Yn2), and (X i,,
Rectangle Wfl determined by X 12) (Y i,, Y i2)
The size of the range is set to RX3. When it is larger than RY3, it is determined that there is a change, and the change area calculation circuit 20 calculates the section exceeding the projection setting value R5 as the change area, and the image of this rectangular area is compressed and encoded as appropriate by the image leaf compression transmission circuit 21. The data is then converted into a message and transmitted to the receiver via a telephone line or the like. Further, the verification data of the changed area projection verification circuit 18 is input to the changed object movement determination and tracking circuit 23 together with the changed object identification data from the changed object matching and identification circuit 16, and the changed object movement determination and tracking circuit 23 moves and tracks the changed object. A stationary bright changing object determined by the light/dark determination circuit 17 is a fire ignition location, a relatively dark changing object, and C is determined to be an intruder even though it is moving. Here, the projection size of the change pixel of the current image of the infrared condyle region obtained by photographing through the bandpass filter 32 in the infrared region is verified, and the size is determined by the bandpass filter 32 in the visible light region.
If the size is larger than the projected size of the changing pixel in the visible light region obtained by photographing through the camera, it is determined as a heat-generating changing object, and in the opposite case, it is determined as a heat-receiving changing object. The intruder determination becomes reliable.

By the way, if you want to send only changed pixels instead of a rectangular area, set the changed pixel determination setting value to R4 (R4<R].
Recalculate the changed pixels and transmit the changed pixel information of the current image.

Detection of signs of illuminance changes, selection of normalization method,
The combinations and parameter settings are optimally set according to the environmental conditions at the monitoring site and the monitored target. These settings are performed by the illuminance change detection method setting circuit 6 and the illuminance change monitoring area setting circuit 27.

Note that Figures 17 and 19 are the current images of Figures 14 and 15,
The buffer image of the binary image of the reference image is shown, and the 21st
The figure shows the buffer image of the difference image. Furthermore, FIGS. 22(a) and 22(b) show the X and Y projections of the difference image. Furthermore, FIGS. 23 to 29 (a) and (b) show the histogram of the current image in FIG. 5 when the monitoring area is defined, the puff 7a image of its binary image, the histogram of the reference image in FIG. 6, It shows the buffer image of the binary image, the buffer image of the difference image between these images, and the X and Y projections of the difference image.

Also, Fig. 31 is the histogram of the image in Fig. 30 with no illuminance change and no intrusion, fJS32 is the buffer image of the binary image, and Fig. 34 is the puff 7 of the binary image of the difference image of the 30th interval. Figures 35(a) and 35(b) show the puff 7a image of the difference image and the X and Y projections of the difference image.

Note that smoothing may be applied to the projection. This is done by calculating the average value of each 0 projection by calculating the average value of the projections before and after the projection, or by weighting the values of the projections before and after the 0 projection. This smoothing can remove small changes.

↓, 'Bottom': : White Figure 3 shows the main circuit configuration of an embodiment that is a modification of the embodiment in Figure 1. In this embodiment, the illuminance normalization calculation circuit 1
3 is stored in the illuminance normalized image frame buffer 29, and the calculation circuit 26 calculates the magnitude of change for each pixel between this normalized image and the current image to obtain a difference image. The difference image is stored in the difference image 7 frame 7a 30, and the pixel change verification circuit 28 replaces the value of the changed pixel with the corresponding pixel value of the current image, and stores it in the changed pixel 7 frame 7a 31. Based on the stored data of the difference image frame buffer 30 and the changed pixel 7 frame 7a 31, the projection calculation circuit 14 selects and calculates projection of the difference image, projection of a binary image of the difference image, or projection of the changed pixel. It was made so that it could be done.

[Effects of the Invention 1] The present invention uses an image monitoring system that detects abnormal changes such as intruders and fire by comparing the current image and a reference image, and uses illuminance changes that detect changes in external light such as sunlight and lighting. a detection means;
The illuminance change detection means may include a means for not detecting an abnormal change in the image when the illuminance change detection output is output, and a means for updating the reference image when the illuminance change is greater than a predetermined value, or a means for comparing the current image and the reference image. to prevent intruders, fires, etc. I! In an image monitoring method that detects constant changes, the current image is compared with a reference image, and if there is a change in the current image, the next image is further compared with the current image to determine the difference between the next image and the current image. If there is a change in the current image, it is determined that there is an abnormal change in the next image and the current image, and if there is no change between the one-way image and the current 1liJ image, the reference image is updated to the current image or the next image or current image is changed. In an image monitoring method that detects abnormal changes such as an intruder or fire by updating the moving average including images or comparing the current image and a reference image, the current lil image and the reference R image are In order to compare, a pixel-by-pixel difference image between the current image and the reference image is obtained, and each pixel of this difference image is defined as the difference between the current image and the reference image, and if the difference is less than or equal to a certain setting value, then When the number of zero pixels is smaller than a predetermined setting value, it is determined that there is a sign of a change in illuminance, and the next image is compared with the current image, and if there is no change, the reference image is updated with the current image, and a change is detected. Since it is determined that there is an abnormality, signs of changes in illuminance can be detected and the reference image can be updated, resulting in the effect that false alarms due to changes in illuminance can be prevented.

[Brief explanation of drawings]

Fig. fjIJ1 is a circuit diagram of an embodiment of the present invention, Fig. 2 is an illuminance normalization calculation circuit same as above, Fig. 3 is a main part circuit diagram of another embodiment of the change detection process, and Fig. 4 is the left side of the TV. The camera arrangement side view and Figures 5 to 35 are explanatory diagrams of the same operation as above, 1 is the TV left camera 7 is an illuminance meter, 8 is an illuminance change detection method setting circuit, 9 is a histogram creation circuit, and 10 is a continuous change This is a judgment circuit. Agent Patent Attorney Man 1) 艮 7th 7th += 9th 8th! 〈 1 o! Fig. 11 Fig. 12 Fig. 13 (b) Fig. 20 1nfer Fig. 21 Fig. 22 is j (b) Fig. 27 Fig. 28 Fig. 29 (b) Fig. 31 (4 intensity Fig. 32 Fig. 33 Fig. 7B34 Figure 35 (b) - Procedural 1] - Positive VF (Method) April 1986: (1 piece

Claims (3)

[Claims] (1) In an image monitoring system that compares the current image and a reference image to detect abnormal changes such as an intruder or fire, an illuminance change detection means that detects changes in external light such as sunlight or lighting; An image monitoring method comprising means for not detecting an abnormal change in an image when the illuminance change detection means outputs an illuminance change detection, and means for updating a reference image when the illuminance change is greater than a predetermined value. (2) In an image monitoring method that compares the current image and a reference image to detect abnormal changes such as an intruder or fire, the current image and the reference image are compared and if there is a change in the current image, ,
Furthermore, the next image and the current image are compared, and if there is a change between the next image and the current image, it is determined that there is an abnormal change in the next image and the current image. An image monitoring method characterized by updating the reference image to the current image or to a moving average including the next image and the current image if the reference image does not exist. (3) In an image monitoring method that compares the current image and a reference image to detect abnormal changes such as an intruder or fire, it is necessary to compare the current image and the reference image. A difference image for each pixel is obtained, and each pixel of this difference image is defined as the magnitude of the difference between the current image and the reference image. An image monitoring method that determines that there is a sign of a change in illuminance, compares the next image with the current image, updates the reference image with the current image if there is no change, and determines that there is an abnormality if there is a change.