JPH06113302A - Detector for person invaded in picture processing - Google Patents

Abstract

PURPOSE:To obtain the detector for person invade in picture processing with excellent reliability without mis-information and missing of information inexpensively by implementing specific algorithm. CONSTITUTION:A 1st processing means 4b obtains a number of picture elements outputted from a picture element processing means 4a for each video image whose contract level is a predetermined contrast level or over and outputs a set output representing the picture elements whose contract level is a predetermined contrast level or over. A 2nd processing means 4c sets an ON output when a maximum difference being a difference between a maximum value and a minimum value of the contrast level of each picture element sequentially outputted from a picture element processing means 4a for each video image is a predetermined difference level or over. A 3rd processing means 4d sets an ON output when number of times of the occurrence of an extreme value of a contrast level of each picture element outputted sequentially from the picture element processing means 4a for each video image is a predetermined extreme occurrence number or over. A notice output section 5 sets an alarm output when the 1st, 2nd and 3rd processing means 4b,4c,4a output an ON-output.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing intruder detecting device for detecting an intruder by processing an image captured by an image pickup camera.

[0002]

2. Description of the Related Art In recent years, an image processing intruder detecting device for detecting an intruder by processing an image captured by an image pickup camera has been developed and put on the market. These image processing intruder detection devices store a so-called standard pattern image in the absence of an intruder in the storage unit, and compare the standard pattern with a new image captured by an imaging camera. The number of changed pixels is counted, and if the count is equal to or greater than a predetermined number, it is determined that there is an intruder. In addition, as a method for determining the presence or absence of an intruder, an image pattern binarized by comparing the standard pattern with an image newly captured by an image pickup camera and judging whether the change exceeds a certain threshold value is stored in advance. If there is a similar pattern that is compared and matched with a similar pattern, it is judged that there is an intruder, or a condition setting function such as changing the size of the detection object or the detection area to be watched is added. Things,
Some have complex algorithm processing.

[0003]

However, in the conventional image processing intruder detection device, as described above, in any case, a so-called intruder-free image called a standard pattern is stored. Therefore, the hardware for storing and recognizing the standard pattern is required, and the hardware and software for renewing the standard pattern are required, which complicates software development and system configuration. There was a problem that it would be very expensive.

The present invention has been made in order to solve the above-mentioned problems, and its purpose is to carry out algorithm processing different from the conventional one, thereby making it inexpensive and reliable without false alarms or loss of information. It is to provide a good image processing intruder detection device.

[0005]

In order to solve the above problems, the present invention provides an image pickup camera, an A / D conversion section, an arithmetic processing section, and an alarm output section. An image processing intruder detection device comprising: an imaging camera for imaging a detection area and outputting a video signal;
The conversion unit digitizes and outputs the video signal from the imaging camera, and the arithmetic processing unit divides one video image from the imaging camera into predetermined pixels based on the output of the A / D conversion unit. Pixel processing means for calculating and sequentially outputting the average brightness level for each pixel, and pixels for which the brightness level of each pixel sequentially output from the pixel processing means for each image is equal to or greater than a predetermined brightness level constant. A first processing means for obtaining a number and turning on the output if the number is equal to or larger than a predetermined pixel constant;
If the maximum difference value, which is the difference between the maximum value and the minimum value of the brightness level of each pixel, which is sequentially output from the pixel processing means for each video image, is greater than or equal to a predetermined difference value constant, the on output is performed.
The processing means and the third processing means for turning on the output when the number of times of occurrence of the extreme value of the brightness level of each pixel, which is sequentially output from the pixel processing means for each image, is equal to or more than a predetermined extreme value generation constant. The alarm output unit is characterized by outputting an alarm if all of the first processing means, the second processing means and the third processing means are ON outputs.

Further, the invention according to claim 2 is characterized in that it is provided with an operating section for setting and changing the light-dark level constant, the pixel constant, the difference value constant and the extreme value generation constant, respectively. Further, the invention according to claim 3 is characterized in that the operation section can specify the pixel to set the valid / invalid.

[0007]

With the above construction, in the invention according to claim 1, the pixel processing means of the arithmetic processing unit is A
Based on the output of the / D conversion unit, one image from the image pickup camera is divided into predetermined pixels, and the average brightness level is calculated for each pixel and sequentially output. Then, the first processing unit of the arithmetic processing unit obtains the number of pixels in which the light-dark level of each pixel sequentially output from the pixel processing unit for each video image is equal to or more than the predetermined light-dark level constant and determines the predetermined pixel constant. If it is above, the output is turned on. The second processing means of the arithmetic processing unit is
If the maximum difference value, which is the difference between the maximum value and the minimum value of the brightness level of each pixel, which is sequentially output from the pixel processing means for each video image, is greater than or equal to a predetermined difference value constant, ON output is performed.
The third processing means of the arithmetic processing section outputs ON when the number of times of occurrence of the extreme value of the brightness level of each pixel, which is sequentially output from the pixel processing means for each image, is equal to or more than a predetermined extreme value generation constant. . Then, the alarm output unit includes the first processing means and the second processing means.
If both the processing means and the third processing means are ON outputs, an alarm is output. Therefore, it is possible to eliminate the conventional storage area for storing the image called the standard pattern, that is, the image when there is no intruder and the software.

According to the second aspect of the present invention, the light-dark level constant, the pixel constant, the difference value constant, and the extreme value generation constant can be optimally changed according to the site.

Further, according to the third aspect of the present invention, the pixel is specified in a scene where an image of changing light and shade which is not an intruder is inevitably partially captured within the image capturing range of the image capturing camera. You can then disable it.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the image processing intruder detecting device according to the present invention will be described in detail below with reference to FIGS. FIG. 1 is a system block diagram for explaining the image processing intruder detection device, and FIG. 2 is an explanatory diagram showing the operation of the arithmetic processing unit of the image processing intruder detection device.

As shown in FIG. 1, the image processing intruder detection apparatus 1 includes an image pickup camera 2, an A / D conversion section 3, and an arithmetic processing section 4.
The alarm output unit 5 and the operation unit 6 are provided. The image pickup camera 2 picks up an image of the detection area A and outputs a video signal. The A / D conversion unit 3 digitizes the analog video signal from the imaging camera 2 and outputs it.

The arithmetic processing section 4 is a section for processing the output from the A / D conversion section 3, and comprises a pixel processing means 4a, a first processing means 4b, a second processing means 4c and a third processing means 4d. Is. The pixel processing unit 4a divides the image from the image pickup camera 2 into predetermined pixels for each image based on the output from the A / D conversion unit 3, and calculates an average brightness level for each pixel. Are sequentially output. Each of the first processing means 4b, the second processing means 4c, and the third processing means 4d processes the brightness level of a series of pixels sequentially output from the pixel processing means 4a for each image. is there. The first processing unit 4b obtains the number of pixels in which the light / dark level of each pixel sequentially output from the pixel processing unit 4a for each image is equal to or more than a predetermined light / dark level constant, and is equal to or more than the predetermined pixel constant. It is designed to output on. The second processing unit 4c has a maximum difference value, which is a difference between the maximum value and the minimum value of the brightness level of each pixel, which is sequentially output from the pixel processing unit 4a for each image, being equal to or greater than a predetermined difference value constant. If there is, it will be turned on. The third processing means 4d turns on the output if the number of times of occurrence of the extreme value of the brightness level of each pixel, which is sequentially output from the pixel processing means 4a for each image, is equal to or more than a predetermined extreme value generation constant. Has been done.

The alarm output unit 5 outputs an alarm if all of the first processing means 4b, the second processing means 4c and the third processing means 4d are ON outputs. As an alarm output,
Relay contact output or alarm ringing sound is used.
The operation unit 6 includes a liquid crystal display unit (not shown) and a ten-key input unit (not shown) so that the light / dark level constant, the pixel constant, the difference value constant, and the extreme value generation constant can be changed. And the pixel processing means 4a
It is possible to specify the pixels of and to change the setting of valid / invalid.

Next, the operation of the image processing intruder detecting apparatus 1 configured as described above will be described with reference to FIG. Figure 2
The pixel processing means 4a divides one image of the detection area A imaged by the imaging camera 2 into predetermined pixels Y _{i, j} based on the output from the A / D conversion unit 3. Each pixel Y
_The figure shows a state in which the average brightness level is calculated for each of _{i and j} .

For convenience of explanation, FIG. 2 shows a case in which one image in the detection area A is divided into pixels Y _{i, j} of a grid matrix of 8 rows and 7 columns. The suffix ij indicates the row number and the column number, respectively. However, the pixel Y _{i, j} is not necessarily limited to the grid matrix of 8 rows and 7 columns, and may be divided into an arbitrary number of pixels in an arbitrary shape. Further, in FIG. 2, each pixel Y _{i, j}
Each number entered in the figure indicates the average brightness level Z _{i, j} of each pixel Y _{i, j} .

The A / D converter 3 digitizes the analog video signal output from the image pickup camera 2 and outputs it. Pixel processing unit 4a, each pixel Y _i, on the basis of the luminance data of the A / D converter 3 corresponding to the coordinates of _j, calculates the average brightness level Z _{i, j} of each pixel Y _{i, j.} In FIG. 2, Z _1,1 =
1, ..., Z _2,3 = 2, ..., Z _2,6 = 1, ..., Z _3,
2 = 2, Z _3,3 = 3, ..., Z _3,6 = 2, Z _3,7 = 1,
_{......, Z 4,2 = 2, Z} 4, 3 = 3, Z 4,4 = 5, Z 4,5 =
3, Z _4,6 = 2, Z _4,7 = 1, ..., Z _5,3 = 3, Z
_5,4 = 4, Z _5,5 = 3, Z _5,6 = 1, ..., Z _6,4 =
2, Z _6,5 = 1, ..., Z _8,2 = 2, _.. , Z _8,7 = 1
Is made.

The pixel processing means 4a of the arithmetic processing section 4 sequentially outputs the series of average brightness levels Z _{i, j} to the first processing means 4b, the second processing means 4c and the third processing means 4d. When the pixel processing means 4a finishes outputting a series of 56 average light and dark level values Z _{i, j} , it outputs a delimiter signal indicating that the output of one image is completed, and A / A again.
Each pixel Y _{i, j of} one new image from the output of the D conversion unit 3
The average brightness level Z _{i, j of} is calculated and sequentially output.

Then, the first processing means 4b of the arithmetic processing section 4
The second processing means 4c and the third processing means 4d form one image.
A series of 5 sequentially output from the pixel processing means 4a for each image
6 pixels Y_{i, j}Brightness level Z_{i, j}Process each
To do. That is, the first processing means 4b has a series of 56
Brightness level Z_{i, j}Brightness level constant C determined in advance in ₀
The number of pixels D above is obtained, and this number of pixels D is predetermined.
Pixel constant C₁If it is above, the output is turned on. For example, in FIG.
, The brightness level constant C₀= 3, the number of pixels D
= 9.

The second processing means 4c obtains the maximum difference value ΔZ _max which is the difference between the maximum value and the minimum value in the series of 56 light and dark levels Z _{i, j} , and this maximum difference value ΔZ _max is previously calculated. If the difference value constant C ₂ is equal to or more than the predetermined difference value constant, the output is turned on. For example, in FIG. 2, the maximum difference value ΔZ _max = (maximum value−
Minimum value) = (Z _4,4 −Z _1,1 ) = (5-1) = 4.

The third processing means 4d has a series of 56 light and dark.
Level Z_{i, j}Permutation of (Z_1,1, ……, Z_1,7, Z_2,1，
……, Z_2,7, Z_3,1, ……, Z_8,7) From the light and dark level
Number Z_{i, j}The number of occurrences E of the maximum value of
An extremum generation constant C that is predetermined by the number E₃If it is above, go out
Force The number of occurrences of this maximum value E is determined by a series of 56 light
Dark level Z_{i, j}In the permutation of, always the previous light and dark level
Le Z_{i, j}To the current brightness level Z_{i, j}When subtracting
Sign of difference value (plus or zero or minor
And the sign of the previous difference value and the current
Compare with the sign of the difference value,
And counting the number of changes from zero to negative
Therefore, it can be easily obtained. For example, in Figure 2
For Z _2,5To Z_2,6When the process of_3,5Or
Z_3,6When the process of _4,4To Z_4,5Processing
When I moved to Z_5,4To Z_5,5When the process of
Z _6,4To Z_6,5When the process of the_8,6
To Z_8,7Occurrence of maximum value when moving to processing
And the number of occurrences E = 6.

The alarm output unit 5 outputs an alarm if the first processing means 4b, the second processing means 4c, and the third processing means 4d are all on outputs. When the series of 56 light / dark levels Z _{i, j} have been processed, the arithmetic processing unit 4 repeats the next series of 56 light / dark levels Z _{i, j} .

The above-mentioned light-dark level constant C₀, Pixel constant
C₁, Difference value constant C₂, Extremum generation constant C ₃Is the operation unit 6
It is possible to change and set each by.
When installing the intruder detection device 1, the actual detection area
When a person invades, the image processing intruder detection device 1 issues an alarm.
Set to a value according to the site while trying whether to apply
To be done.

The operation unit 6 is a special feature of the pixel processing means 4a.
Fixed pixel Y_{i, j}Brightness level Z_{i, j}Pixel Y
_{i, j}The settings can be changed for each. That is,
By the operation unit 6, for example, the pixels in the first column in FIG.
(Y_1,1, Y_2,1, Y_3,1, Y _4,1, Y_5,1, Y_6,1，
Y_7,1, Y_8,1) Brightness level (Z_1,1, Z_2,1, Z_3,
1, Z_4,1, Z_5,1, Z_6,1, Z_7,1, Z_8,1) Processing
You can also set to skip (disable)
I'm told. This function installs the imaging camera 2
If the lighter and darker are not the only intruders, for example
For example, the fluctuation of the leaves of the foliage plant was captured as a video.
It is effective when visiting.

[0024]

Since the image processing intruder detecting device of the present invention is constructed as described above, the invention according to claim 1 is a so-called intruder, which is a conventional standard pattern. Since there is no need to store the video when there is no time, and the time to read the comparison data from the standard pattern can be reduced, the storage capacity is small and the cost is low, and the number of images processed per image per unit time. Can be overlooked even if the number of pixels increases and the speed of intrusion is rapid, and the intruder is detected using three parameters: the number of pixels that are equal to or greater than the light-dark level constant, the maximum difference value, and the number of times the extreme value occurs. Therefore, the reliability can be improved.
According to the invention described in the above, in addition to the above effects, the condition setting of the above-mentioned parameters can be changed according to the situation of the site, so that the reliability can be further improved. In addition to the above-described effects, it is possible to provide an image processing intruder detection device that is excellent in usability and can be provided with a certain degree of flexibility in installing an image pickup camera in the field.

[Brief description of drawings]

FIG. 1 is a system block diagram for explaining an image processing intruder detection device according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing the operation of the arithmetic processing unit of the image processing intruder detection apparatus according to the embodiment of the present invention.

[Explanation of symbols]

1 Image Processing Intruder Detection Device 2 Imaging Camera 3 A / D Converter 4 Arithmetic Processing Unit 4a Pixel Processing Means 4b First Processing Means 4c Second Processing Means 4d Third Processing Means 5 Alarm Output Unit 6 Operation Unit A Detection Area Y _{i, j} pixels

Claims (3)

[Claims] 1. An image processing intruder detection device comprising an imaging camera, an A / D conversion unit, an arithmetic processing unit, and a warning output unit, wherein the imaging camera images a detection region and outputs a video signal. The A / D conversion unit digitizes and outputs the video signal from the imaging camera, and the arithmetic processing unit outputs 1 from the imaging camera based on the output of the A / D conversion unit.
Pixel processing means for dividing a video image into predetermined pixels, calculating an average brightness level for each pixel, and sequentially outputting the calculated levels.
First processing means for obtaining the number of pixels in which the brightness level of each pixel sequentially output from the pixel processing means for each video image is equal to or greater than a predetermined lightness / darkness level constant, and if the number of pixels is equal to or more than the predetermined pixel constant, performs ON output. The second process of turning on the output if the maximum difference value, which is the difference between the maximum value and the minimum value of the brightness level of each pixel, which is sequentially output from the pixel processing means for each video image, is greater than or equal to a predetermined difference value constant. And a third processing means for performing on-output if the number of times of occurrence of the extreme value of the brightness level of each pixel sequentially output from the pixel processing means for each image is equal to or more than a predetermined extreme value generation constant. An image processing intruder detection device, wherein the alarm output unit outputs an alarm if all of the first processing means, the second processing means, and the third processing means are ON outputs. 2. The image processing intruder detection device according to claim 1, further comprising an operation unit capable of changing settings of the light-dark level constant, the pixel constant, the difference value constant, and the extreme value generation constant. 3. The image processing intruder detection device according to claim 2, wherein the operation unit specifies a pixel to enable / disable the pixel.