JPH0397080A - Picture monitoring device - Google Patents

Abstract

PURPOSE:To prevent the detection of an intruder from being erroneously transmitted by counting the number of continuous picture element trains as the luminance difference of each picture calculated from input picture information and reference picture information, deciding continuity and generating an alarm when the number of the continuous picture element trains is more than the prescribed continuity. CONSTITUTION:An input from a picture input device in a specified monitoring area is executed as the set information of plural picture elements and these information are compared with the previously acquired picture information of a reference memory 3. Then, it is discriminated from a difference value to a reference picture that any abnormality is generated in the speci fied monitoring area. Further, the information are converted to binary information and after wards, the continuity of '1' data is discriminated in the longitudinal and lateral picture element trains. In comparison with the threshold value of the continuity based on disturbance causes, it is discriminated whether the picture is changed by the disturbance cause or by the real existence of the intruder. When the real intruder is discriminated, an alarm generating circuit 5 is driven. Thus, even when a TV camera or the specified monitoring area is vibrated by wind, the passage of a car or the construction of a road, etc., it can be prevented that the alarm operation is executed by erroneous judgement.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention monitors intruders in a pre-specified area to be monitored (hereinafter referred to as a specific monitoring area) based on image information obtained by imaging. Regarding crime prevention monitoring devices that issue alarms, in more detail, changes in image information due to intruders into the monitoring area, and image information due to vibrations of the imaging camera due to some cause or vibrations in a specific monitoring area that cause false alarms. This paper relates to an image monitoring device that realizes highly accurate indigo vision by reliably identifying changes in color. [Prior art] Conventionally, a specific monitoring area is imaged with a TV camera, etc., and this image information is analyzed to determine if a person enters the monitoring area at a time when it is not the time when people should enter. An image monitoring device that detects an intruder and issues an alarm has been proposed and put into practice. These image monitoring devices compare image information taken sequentially with a TV camera, etc., calculate changes in the image, that is, changes in the brightness of the image, and if a change in brightness exceeds a certain value, the system automatically monitors the same specific area. This is the principle that determines that there is a human intrusion, that is, an intruder.

[Problem to be solved by the invention]

However, in the conventional image monitoring device described above, for example,
Vibrations caused by strong winds or when a large vehicle passes by may cause the imaging TV cameras installed indoors in the building that constitutes the specific monitoring area to vibrate, or the building itself that has the monitoring area may be affected by an earthquake. When vibrations occur due to road construction or the like, the brightness of the image naturally changes, causing a problem in that the user may mistakenly judge that there is an intruder even though there is no intruder.

SUMMARY OF THE INVENTION Therefore, in view of the above-mentioned problems, an object of the present invention is to prevent erroneous judgments and alarms even when a TV camera or a specific monitoring area vibrates due to wind, passing cars, road construction, etc. The objective is to provide an image monitoring device that has been improved and devised so that it does not operate.

[Solution]

That is, the present invention inputs an image of a specific monitoring area as aggregate information of a plurality of pixels, obtains the luminance of each pixel using a plurality of gradations, and compares this with reference image information obtained in advance. First, it is determined that some kind of difference has occurred in the specific monitoring area from the difference value with respect to the reference image, and then this difference value is added for each vertical and horizontal pixel column, and then "0" and "1" are determined. After converting to binary information, "1" in the vertical and horizontal pixel rows is
” By determining the continuity of the data and comparing it with the continuity threshold based on the disturbance factor, it is possible to determine whether the image change is due to the disturbance factor or the image change is actually caused by the presence of an intruder. The latter is designed to activate an alarm generation circuit when an actual intruder is identified. According to the present invention, in an image monitoring device that detects changes in image information in a specific monitoring area and monitors for intruders,
a reference image memory that stores a reference image of a specific monitoring area; an image processing device that captures and inputs images of the specific monitoring area; and an image processing device that determines the presence or absence of an intruder from the manually generated image information and the reference image information. The image processing device includes pixel column adding means for adding, for each image column, the luminance difference of each image calculated from the input image information and reference image information, and each pixel column adding means added by the pixel column adding means. and continuity determination means for counting the number of consecutive pixel rows whose differential luminance value is equal to or greater than a certain value, and configured to issue an alarm when the determination result of the continuity determination means exceeds a predetermined value. An image monitoring device is provided.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS The image monitoring apparatus according to the present invention will be described in more detail below based on embodiments shown in the accompanying drawings.

FIG. 1 shows a functional block diagram of an image monitoring device according to an embodiment of the present invention.

The image monitoring device shown in FIG. 1 has an imaging capability that covers a prespecified monitoring area, that is, a specific monitoring area,
It is equipped with an image input device l configured with a camera or the like, and constantly captures and inputs image information of the specific monitoring area. The image information captured by the image input device 1 and manually input is sequentially stored in the input image memory 2, but this input image is (nXm)
It is stored as a two-dimensional information group formed from individual pixels.

Further, this image monitoring device uses a reference image with which the input image of the input image memory 2 is compared, for example, image information of the same specific monitoring area in a completely unmanned state as an image for comparison reference.
The input image memory 2 and the reference image memory 3 are used in the next stage image processing device 4.
The image processing device 4 consists of an arithmetic processing device constituted by a well-known microprocessor means, etc., and includes an absolute difference calculation means 4l, a pixel string addition means 42, a binarization means 43, and a continuity determination means. It is equipped with means 44 and intruder determining means 45. This image processing device 4 further includes:
An alarm device 5 is connected. The absolute difference calculation means 4l of the image processing device 4 is connected to the input image memory 2 and the reference image memo I7-3, and calculates the absolute value of the difference in brightness for each pixel forming image information. The pixel column addition means 42 is a means for summing the difference luminance of each pixel calculated by the absolute difference calculation means 41 for each pixel column in the horizontal and vertical directions of the two-dimensional image. When the next stage binarization means 43 receives the luminance added at the pixel column addition stage 42,
This brightness value is selected in advance and compared with a certain brightness value t+S'' that forms the input darkness value, and if it exceeds the darkness value, it is binarized as "1", and if it is less than that value, it is binarized as rQJ. Continuity determining means 44 connected to this binarizing means 43 determines the continuity of each value (0 or 1) in the horizontal and vertical directions binarized by the binarizing means 44, that is, 1 is the same. This is a means for determining how many 1's appear consecutively in a row or column, and is determined by counting the number of consecutive 1's. The intruder determining means 45 compares the count value of the continuity determining means 44 with a predetermined value R, determines the presence or absence of an abnormality, that is, the presence or absence of an intruder in the specific monitoring area, and sends the determination result to the subsequent alarm device 5. Output. Next, the operation of the image monitoring device having the above-mentioned basic structure will be explained below.

To explain the processing procedure of the image monitoring device shown in FIG. 1 according to the flowchart shown in FIG. An image of the specific monitoring area is captured by the image input device 1 (301) and input to the reference image memory 3 (302). The image captured and input here is as shown in FIG. ~A. f
It is composed of i pixels (for example, 512 x 512), and each pixel has, for example, 256 levels of brightness from 0 to 255. are doing.

Next, when the image monitoring device receives an activation signal based on a warning setting controller (not shown), the device enters the monitoring system. Image the specific monitoring area with the human-powered image device 1 (SO3),
The image is temporarily stored in the human image memory 2 (S04). The image information to be stored is B.Same as the reference image above. It is composed of pixels of ~B■.

At this time, each time a trigger pulse is emitted by a crystal oscillator (not shown), captured images of the specific monitoring area N are sequentially manually inputted into the manually operated image memory 2 .

The absolute difference calculation means 41 takes in images from both the reference image memory 3 and the input image memory 2, and calculates the difference between the brightness values of the pixels corresponding to each of the two pieces of image information.
The absolute value of the difference C. ~C■ is now calculated and temporarily stored in buffer memory means (not shown). Note that C++=I
A++ B. l,cl! =lAI! 1B. ．． l, -
・・・- ・・C. ．． = IA,,B■1.

? Then, the pixel column addition means 42 calculates the calculated brightness difference absolute value C. from the unillustrated Batza memory. ~
C■ is read out, and the absolute difference value of each pixel is added column by column to the pixel columns in the horizontal and vertical directions of the image (306
).

As shown in FIG. 4, the horizontal row is Ht=Cth+C+
Z+...+Clfi, H2=C■+C2■+...
・・・・・102fi、・・・・・・H. =C■+C. !
+・・・+C■, and the vertical column is ■1 =C
．． +Ct. +...+C■, Vz=CIz10C'zz
+ ...+ Cv-z, """' Vll=
C+. Ctn+...10■.

As mentioned above, H, ~H. , V. After calculating ~■7, these H1~HII, vI~■7 are temporarily stored in a buffer memory (not shown).

Next, the binarization means 43 converts }{ l ””' H II
1, and ■, to ■7 are sequentially read from the memory and compared with a threshold value S (for example, S = 2,000), and the values greater than or equal to the value of S are binarized as 1, and those less than S are binarized as O (SO7), Fifth
As shown in the figure, HB. -HB,,VB, ~VB. Suppose that As a result, if the input image has changed greatly compared to the brightness of the reference image, that is, if there is an intruder in the image, the pixel row containing the intruder will be "1", and the pixels containing other pixels will be "1". The column will be "0".

The continuity determining means 44 performs the above-mentioned HB, to HB. and VB
,~VB. , each consecutive number of "1" is counted (SOS). Based on the consecutive number count result of the continuity determining means 44, the intruder determining means 45 determines a predetermined value R(
For example, R=5), and if both the vertical and horizontal directions are R or less, it is a change in the image due to vibration of the TV camera or the monitoring area, and the monitoring system returns again. Then, by inputting a new monitoring image from the image input means 1 and updating the input image memory 2, the above-described operation process is repeated.

If there is a series of "1"s equal to or higher than R in either the vertical direction or the horizontal direction, it is determined that the image change is not caused by a disturbance and that it is an intruder, and that fact is input to the alarm device 5 (310).

Here, in the above explanation, the vertical and horizontal thresholds were set to the same value, but they may be different values, and by setting this darkness value, it is possible to distinguish between cases where a small animal has invaded and cases where a human has invaded. It is also possible to differentiate.

In addition, in the above explanation, rl in both the vertical and horizontal directions is
Although the number of consecutive J's is determined, it may be determined only in the vertical direction or only in the horizontal direction depending on the shape of the intruder to be monitored.

It should be noted that the alarm device 5 is a well-known means for notifying the outside by, for example, lighting an alarm lamp (not shown) or sounding a buzzer if there is an input from the intruder determining means 45. Here, the principle of determination by the continuity determining means 44 will be explained with reference to FIGS. 6 to 8.

FIG. 6 shows the reference image mentioned above. On the other hand, if there is a moving object such as an intruder in the input image from the image input device 1, as shown in FIG. 7, the image portion whose brightness has greatly changed from the reference image covers a relatively large range. Therefore, in the information corresponding to this image part, HB, ~
H.B. In the binary information of , VB, ~VB, "1" is consecutive.

On the other hand, when vibrations occur in the image input device 1 or in the entire specific monitoring area, as shown in FIG.
The range in which the brightness changes significantly is small, and therefore the number of consecutive rlJs is also small. Therefore, this predetermined consecutive number is determined in advance by taking into consideration the image changes caused by disturbance factors, the installation situation, etc. Based on this fact, the continuity determining means 44 can clearly distinguish between the two by comparing them with the fixed threshold value R.

〔Effect of the invention〕

As can be understood through the embodiments described above, according to the image monitoring device according to the present invention, a TV that images a specific monitoring area
Even if the camera or the specific monitoring area itself vibrates due to exceeding its stable operating state due to strong winds or the passage of a large vehicle, and this causes changes in the input image of the specific monitoring area, an intruder may not be able to capture the image. Since changes can be reliably distinguished, malfunctions of the monitoring device, that is, erroneous transmission of intruder detection can be prevented. Further, it is possible to prevent unnecessary or erroneous responses by personnel when an abnormality is detected, and it is possible to reduce costs in the security system.

[Brief explanation of drawings]

FIG. 1 is a functional block diagram of an image monitoring device according to an embodiment of the present invention, FIG. 2 is a flowchart explaining the operation of the image monitoring device, and FIG. 3 is a two-dimensional diagram of a plurality of pixels showing image information. FIG. 4 is a plan view showing the arrangement; FIG. 4 is a conceptual diagram showing the results of calculating the difference values between the reference image information of the specific monitoring area and the input image information in vertical and horizontal columns; and FIG. 5 is the binarization means. A conceptual diagram showing the result of binarizing the difference value in Fig. 4 using The figure which shows the image of a disturbance factor, etc. ■...Image input device, 2...Input image memory 3.
...Reference image memory, 5...Alarm device. 4... Image processing device,

Claims (1)

[Scope of Claims] 1. An image monitoring device that detects changes in image information in a monitoring area and monitors intruders, etc., comprising: a reference image memory that stores a reference image of a specific monitoring area; and the specific monitoring area. an image processing device that determines the presence or absence of an intruder based on the input image information and the reference image information; pixel column adding means for adding the luminance differences of each image calculated from each pixel column, and continuity for counting the number of consecutive pixel columns in which each difference luminance value added by the pixel column adding means is equal to or greater than a certain value. 1. An image monitoring device comprising: a determining means, and generating an alarm drive signal to an alarm circuit when the number of consecutive determination results of the continuity determining means is equal to or greater than a predetermined number of consecutive times. 2. The image processing device further includes a binarization unit that binarizes the luminance difference value of each image added by the pixel string addition unit according to the magnitude with respect to a certain threshold value, and the binarization unit binarizes the luminance difference value of each image added by the pixel string addition unit. 2. The image monitoring apparatus according to claim 1, wherein the continuity determining means counts the number of consecutive pixel rows whose differential luminance values are equal to or greater than the predetermined value.