JP2002247567A - Monitoring camera device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To quickly and appropriately cope with the situation that a monitoring camera is in a monitoring impossible state. SOLUTION: By an image fetch part 3, background images from a camera 1 are written to a background image memory 4 and monitoring images are written to an image A memory. They are read, the luminance difference of the background image and the image A is processed in a luminance difference processing part 7 and binarized in a binarization processing part 8 and an abnormal area is extracted in an area extraction part 9. When the abnormal area is larger than a threshold, an auxiliary illumination 11 is lighted, the image at that time is written to an image B memory and read in the luminance difference processing part and the luminance difference of the image A and the image B is processed and binarized. When the abnormal area is smaller than the threshold, luminance distribution is calculated in a luminance distribution calculation part 12. When it is larger than a distribution value threshold, the interruption of video signals from the camera is decided in a decision processing part 10. When it is smaller, it is decided that only the brightness of a part of the images is changed. When the abnormal area is larger than the threshold, the luminance difference of (background image-image A)-(image A-image B) is obtained, a cover is put on the camera when the abnormal area is larger than the threshold and it is decided that the illumination is turned off when it is smaller.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a surveillance camera device and, more particularly, to a surveillance camera device for determining a cause of an inability to monitor when a monitoring failure occurs.

[0002]

2. Description of the Related Art When a surveillance camera (hereinafter abbreviated as "camera") falls into an unmonitoring state and it becomes impossible to monitor an intruder or to monitor an article, a coping method differs depending on the cause. is important. One of the reasons is that the camera may be covered with something.
Conventionally, auxiliary lighting is provided near the camera, the auxiliary lighting is turned on, and the light is reflected by the cover, so that it is possible to determine that the camera is covered by the state of increase in the average brightness of the screen due to the reflected light. There was something. However, for example, when the front of the camera lens is covered, the brightness of the screen does not increase even if the auxiliary lighting is turned on. Cannot be determined.

[0003]

SUMMARY OF THE INVENTION It is an object of the present invention to determine the cause of an unmonitored state on the basis of a video signal from the camera when the camera falls into an unmonitored state, and to promptly deal with it. And

[0004]

To achieve the above object, a surveillance camera device according to the present invention comprises a camera, an auxiliary lighting provided near the camera, and a background image when there is no abnormality from the camera. A background image memory for storing, an image A memory for storing a monitoring image sequentially output from the camera, an image B memory for storing an image from the camera when the auxiliary lighting is turned on, the background image memory and the image A
First luminance difference of image read from memory, or image A
A luminance difference processing unit that processes a second luminance difference of the image read from the memory and the image B memory, or a luminance difference between the first luminance difference and the second luminance difference, and binarizes data from the luminance difference processing unit A binarization processing unit, and an area extraction unit that extracts an abnormal area based on data from the binarization processing unit;
A determination processing unit that determines whether there is an abnormality in the monitoring image based on data from the area extraction unit, or whether the state of the monitoring camera device is normal, the background image memory, the image A memory,
It comprises an image B memory, a luminance difference processing section, a binarization processing section, an area extraction section, and a control section for controlling the determination processing section.

If the abnormal area of the first luminance difference processed by the luminance difference processing section and the binarization processing section and extracted by the area extraction section is larger than a predetermined area threshold, monitoring by the determination processing section is impossible. It is determined that it is in the state.

When the abnormal region of the first luminance difference is larger than a predetermined region threshold, the auxiliary illumination is turned on, the luminance difference processing unit processes the second luminance difference, and the binarization processing unit The area extracting unit extracts an abnormal area having the second luminance difference. If the extracted abnormal area is smaller than a predetermined threshold, the determination processing unit interrupts the video signal from the camera. Or that the brightness of a part of the image has changed.

[0007] A luminance variance calculator for quantifying the luminance distribution of the image is provided, and the second variance calculator extracted by the area extractor is used.
When the abnormal region of the luminance difference is smaller than the predetermined region threshold, the luminance variance of the abnormal region is calculated. When the luminance variance value is smaller than the predetermined variance threshold value, the video signal from the monitoring camera is obtained. Is determined to have been interrupted, and if it is greater than the variance threshold, it is determined that the brightness of a part of the image has changed.

If the abnormal area of the second luminance difference extracted by the area extracting unit is larger than a predetermined threshold, whether the camera is covered by the determination processing unit,
Alternatively, it is determined that the lighting is off.

When the abnormal region of the second luminance difference extracted by the region extracting unit is larger than a predetermined threshold, the luminance difference processing unit processes the luminance difference between the first luminance difference and the second luminance difference. Processing in the binarization processing unit, extraction of an abnormal region in the region extraction unit, if the extracted abnormal region is larger than a predetermined threshold, the camera is covered, the threshold If it is smaller, it is determined that the lighting is off.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described based on embodiments with reference to the drawings. FIG. 1 is a block diagram of a main part of an embodiment of a surveillance camera device according to the present invention. In FIG. 1, 1 is a camera, 2 is an A / D converter for converting a video signal from the camera 1 into a digital signal, and 3 is an A / D converter. An image capturing unit that captures an image from the D conversion unit 2 into the background image memory 4, the image A memory 5, or the image B memory 6, and the background image memory 4 stores a background image when there is no abnormality in the monitoring field of view of the camera 1. And image A
The memory 5 stores the monitoring images sequentially output from the camera 1, the image B memory 6 stores the image from the camera 1 when the auxiliary lighting 11 is turned on, and the memory 7 stores the images from the background image memory 4 and the image A memory 5. The first luminance difference of the read image, or the second luminance difference of the image read from the image A memory 5 and the image B memory 6, or the first luminance difference and the second luminance difference
A luminance difference processing unit that processes the luminance difference of the luminance difference, 8 is a binarization processing unit that binarizes the data from the luminance difference processing unit 7, and 9 is an abnormal area based on the data from the binarization processing unit 8. An area extraction unit 10 to be extracted is a determination processing unit that determines whether there is an abnormality in the monitoring image based on data from the area extraction unit 9 or whether the state of the camera 1 is normal.
And a light-emitting diode or the like as a light source. Reference numeral 12 denotes a luminance variance calculating unit for quantifying the luminance distribution of an image, and an abnormal region of the second luminance difference extracted by the region extracting unit 9 is a predetermined region. 13 is for calculating the variance of the luminance of the abnormal area when the threshold value is smaller than the threshold, 13 is an intrusion notifying section for notifying that there is an intruder in the judgment processing section 10, and 14 is a judgment processing section.
An abnormality reporting unit that reports an abnormality of the camera 1 when it is determined that there is an abnormality in the camera 10 and a control unit 15 that controls each unit.

Next, the operation of the surveillance camera device according to the present invention will be described with reference to the flowchart of FIG. 2 and various image diagrams of FIG. When there is no abnormality in the monitoring field of view of camera 1,
The video signal of the background image (reference image) from the camera 1 is A /
The digital signal is converted into a digital signal by the D conversion unit 2, and is written into the background image memory 4 by the image capturing unit 3 (step 1, hereinafter referred to as ST).
1). FIG. 3A is an example of a background image, in which an object 22 is placed on a floor 21. Then, the image signal sequentially output from the camera 1 and digitally converted by the A / D conversion unit 2 is written into the image A memory 5 by the image capturing unit 3, and the background image is read from the background image memory 4 by the luminance difference processing unit 7. , An image (image A) is read from the image A memory, and a luminance difference (first luminance difference) for each pixel of the background image and the image A is processed (ST2), and input to the binarization processing unit 8;
The data (difference) of each pixel is binarized into 1 and 0 by comparing it with a predetermined reference value, an abnormal area (area of 1) is extracted by the area extracting section 9 (ST3), and input to the determination processing section 10. . The determination processing unit 10 compares the input data with an area threshold for determining whether the camera is an intruder or an abnormality of the camera 1, and determines that the input data is smaller than the area threshold, that is, a partial change of the monitoring image.
(ST4: No), it is determined that there is an intruder (ST5), and the intrusion notification unit
When a predetermined notification is performed via 13 and it is larger than the area threshold
(ST4: Yes) determines that the camera 1 has some abnormality.

Then, the auxiliary lighting 11 is turned on via the control unit 15 (ST6), and the image from the camera 1 at this time (hereinafter referred to as image B) is written into the image B memory 6 by the image capturing unit 3, and the luminance difference The processing unit 7 reads the image A from the image A memory 5 and the image B from the image B memory 6, and performs processing of a luminance difference (second luminance difference) for each pixel between the image A and the image B (ST).
7) The binarization processing section 8 compares the luminance difference of each pixel with a predetermined reference value to binarize it into 1 and 0, and extracts an abnormal area (area 1) with the area extracting section 9 (ST8). Is input to the determination processing unit 10 and compared with a predetermined area threshold value, and is smaller than the area threshold value (when the auxiliary illumination is turned on (image B) as shown in FIG. 3B or FIG. 3C and FIG. When there is no difference in time (image A) (ST
9: No), the data from the region extraction unit 9 is input to the luminance variance calculation unit 12 via the control unit 15, and the luminance variance calculation for quantifying the brightness distribution of the image by statistical processing is performed (ST10). If the calculated value is smaller than the variance threshold (ST11: No), ie, FIG.
As shown in (b), when the screen is completely dark regardless of the presence or absence of the auxiliary lighting 11, the determination processing unit 10 determines that there is no video signal output from the camera 1 and that the camera 1 may have failed (ST12). ), The abnormality reporting unit 14 reports that fact. If the value calculated by the luminance variance calculator 12 is larger than the variance threshold (ST11: Yes), the auxiliary lighting 11 is turned on as shown in FIG.
There is no change in the distribution of brightness of the image depending on the presence or absence of the camera.
It is determined that the failure is not a failure (ST13), and the abnormality reporting unit 14 reports that fact.

If the abnormal region is larger than the region threshold in ST9 (ST9: Yes), the luminance difference processing section 7 converts the first luminance difference (background image-image A) to the second luminance difference (image A-image B).
Is subtracted (ST14), the binarization processing unit 8 compares the luminance difference of each pixel with a predetermined reference value to binarize it into 1 and 0, and extracts the abnormal region (region 1) with the region extraction unit 9. (ST15), the judgment processing unit
10 and compared with a predetermined area threshold value. If the area value is larger than the area threshold value (ST16: Yes), as shown in FIG. ,
Since the screen is partially brightened by the lighting of the auxiliary lighting 11 as shown in FIG. 7F, it is determined that the camera 1 has been covered in front of the camera 1 by human or natural mischief (ST17), The reporting unit 14 reports that fact. If the abnormal region is smaller than the region threshold in ST16 (ST16: No), the image becomes slightly brighter as shown in FIG. It is determined that the lighting has been turned off (ST18), and the abnormality reporting unit 14 reports that fact.

[0014]

As described above, according to the surveillance camera device according to the present invention, when the surveillance camera falls into a state in which monitoring is not possible, the cause is correctly determined, so that quick and appropriate measures can be taken. It is possible to solve the problem that the monitoring cannot be performed for a long time and the monitoring cannot be performed for a long time.

[Brief description of the drawings]

FIG. 1 is a main block diagram of an embodiment of a monitoring camera device according to the present invention.

FIG. 2 is a flowchart for explaining the operation of the surveillance camera device according to the present invention.

FIG. 3 is an example of images from a camera at normal times and at times of various abnormalities.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 camera 2 A / D conversion unit 3 image capture unit 4 background image memory 5 image A memory 6 image B memory 7 luminance difference processing unit 8 binarization processing unit 9 area extraction unit 10 judgment processing unit 11 auxiliary lighting 12 luminance dispersion Calculation unit 13 Intrusion notification unit 14 Abnormality notification unit 15 Control unit 21 Floor 22 Object

Claims (6)

[Claims] 1. A surveillance camera, an auxiliary lighting provided near the surveillance camera, a background image memory for storing a background image when there is no abnormality from the surveillance camera, and a surveillance sequentially output from the surveillance camera An image A memory for storing an image, an image B memory for storing an image from the surveillance camera when the auxiliary lighting is turned on, a first luminance difference between the background image memory and the image read from the image A memory, or an image A Of the image read from the memory and the image B memory
A luminance difference processing unit that processes a luminance difference or a luminance difference between the first luminance difference and the second luminance difference; a binarization processing unit that binarizes data from the luminance difference processing unit; An area extracting unit that extracts an abnormal area based on data from the processing unit, and a determination processing unit that determines whether there is an abnormality in the monitoring image or whether the state of the monitoring camera device is normal based on data from the area extracting unit And a control unit for controlling the background image memory, the image A memory, the image B memory, the luminance difference processing unit, the binarization processing unit, the area extraction unit, and the determination processing unit. 2. When the abnormal region of the first luminance difference processed by the luminance difference processing unit and the binarization processing unit and extracted by the region extraction unit is larger than a predetermined region threshold, the determination processing unit 2. The surveillance camera device according to claim 1, wherein the surveillance camera device determines that the surveillance is impossible. 3. When the abnormal area of the first luminance difference is larger than a predetermined area threshold, the auxiliary lighting is turned on, and the second luminance difference is processed by the luminance difference processing unit. The processing unit performs processing, and the region extracting unit extracts an abnormal region having the second luminance difference. If the extracted abnormal region is smaller than a predetermined threshold value, the determination processing unit performs image processing from the surveillance camera. 3. The surveillance camera device according to claim 2, wherein it is determined that the signal has been interrupted or that the brightness of a part of the image has changed. 4. A luminance variance calculation unit for quantifying a luminance distribution of an image, wherein an abnormal region of the second luminance difference extracted by the region extracting unit is smaller than a predetermined region threshold. Performs a brightness variance calculation, and the determination processing unit determines that the video signal from the monitoring camera is interrupted when the brightness variance value is smaller than a predetermined variance value threshold; 4. The surveillance camera device according to claim 3, wherein it is determined that a part of the brightness has changed. 5. When the abnormal region of the second luminance difference extracted by the region extracting unit is larger than a predetermined threshold, the monitoring processing unit is covered by the determination processing unit or the illumination is turned off. 4. The surveillance camera device according to claim 3, wherein the surveillance camera device determines that the monitoring has been performed. 6. When the abnormal region of the second luminance difference extracted by the region extracting unit is larger than a predetermined threshold, the luminance difference processing unit processes the luminance difference between the first luminance difference and the second luminance difference. Is performed, the binarization processing unit performs processing, and the area extraction unit extracts an abnormal area. If the extracted abnormal area is larger than a predetermined threshold, the surveillance camera is covered. The surveillance camera device according to claim 5, wherein the determination is made, and if the threshold value is smaller than the threshold value, it is determined that the illumination is turned off.