JP2000341677A - Image monitoring device - Google Patents

Abstract

(57) [Summary] An image monitoring apparatus for detecting an intruder in a monitoring area eliminates erroneous detection due to light or shadow and enables accurate detection of the intruder. SOLUTION: In an image obtained by photographing a monitoring area, an area having a very high luminance is detected as a light area 24. When the change area 19 of the difference between the reference image 11 and the current image 16 includes or includes a light area, it is determined that light has been detected, and the influence is removed. If the change area is in contact with the light area in the light area and has a lower luminance than the light area, it is determined that a shadow has been detected and the influence is removed. As a result, light or shadow in the captured image is not detected as an intruder, and false alarm can be prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image monitoring apparatus for detecting an intruder in a monitoring area by capturing the monitoring area and processing the captured image.

[0002]

2. Description of the Related Art Conventionally, there is an image monitoring apparatus having a function of detecting an intruder in a monitoring area. In this image monitoring device, the monitoring area is continuously photographed by the imaging means,
The current image of the latest situation taken by the image processing means,
By comparing with a reference image stored in advance, a region where there is a change between both images is extracted. When the change area is larger than a predetermined size and moves a predetermined distance or more, it is determined that the change area between the images is caused by an intruder in the monitoring area.

[0003] When detecting an intruder, the image monitoring device memorizes that an intrusion abnormality has occurred as security state information,
An image at the time of occurrence of an abnormality (abnormal image) is stored. When there is a call from the outside such as a controller or a control center, the image monitoring device outputs the stored state information and the abnormal image to the outside. A specific example of the intruder detection logic in the image monitoring device will be described with reference to FIG.

FIG. 1A shows a reference image 11 which has been photographed in advance of a monitoring area by an imaging means and stored in a memory. The reference image 11 includes a floor 12, a side wall 13, and a front wall 1.
4 and window 15 are shown. (B) shows the latest image (current image) 16 taken by the imaging means, in which an intruder 17 is shown. Further, the current image 16 in (B) is an image when the intruder 17 first appears in the monitoring area (first appearance).
(C) is a difference image 18 obtained by taking a difference for each pixel between the reference image 11 and the current image 16, and the intruder 17 is extracted as a change area 19. The position of the change area 19 is defined as a first appearance position 20.

[0005] The image monitoring apparatus captures a monitoring area at intervals of, for example, 0.5 seconds to obtain a current image. (D)
This is the current image 16 captured after a lapse of time from the current image 16 in (B), in which the intruder 17 has moved.
(E) shows a difference image 18 between the reference image 11 and the current image 16. The difference between the current position 22 of the current change area 19 and the first appearance position 20 is the moving distance 23.

In the difference image 18 shown in FIG.
When the object 7 is in the monitoring area, the changing area 19 appears in the current image 16 with a size equal to or larger than a predetermined size, and the changing area 19 moves as the intruder 17 moves. Therefore, the image processing means of the image monitoring device determines that the intruder 17 is in the monitoring area when the size of the change area 19 is within the predetermined range and the moving distance 23 is equal to or more than the predetermined value.

In the above-described intruder detection logic, when light enters the monitoring area, the portion into which the light is inserted becomes a high-luminance area (hereinafter, referred to as "light area") on the captured image. When this light area is generated or disappears, the changed area 19 is detected and detected as an intruder. FIG. 2 is a diagram illustrating the reason why the light area is erroneously determined as an intruder.

FIG. 2A shows a reference image 11,
(B) represents the current image 16, and (C) is a difference image 18 between the reference image 11 and the current image 16. The current image 16 has a window 5
There is a light area 24 generated when sunlight enters from the area. When the reference image 11 and the current image 16 are compared, the light area 24 is detected as the change area 19 in the difference image 18. In addition, when the sunlight has disappeared from the state in which the sunlight has entered, the change region 19 is similarly detected. In the conventional image monitoring device, if the change area 19 moves with the movement of the sun,
This will be detected as an intruder.

When a person passes outside the monitoring area and is projected onto the light area 24, and a shadow moves inside the light area 24, the shadow is also detected as a moving area that moves. Therefore,
Also in this case, the shadow is detected as an intruder. As described above, when light or a shadow is detected as an intruder, the image monitoring apparatus erroneously reports to the controller or control center every time light is inserted or a person passes outside the monitoring area. On the other hand, in the conventional image monitoring apparatus, it has been practiced to distinguish these lights or shadows from intruders and prevent false reports.

If the incident light is faint light, a floor pattern appears in the light area 24 as shown in the current image 16 of FIG. Therefore, the correlation in the change area 19 between the reference image 11 and the current image 16 is high, or the edge information is stored. Therefore, in the related art, an intruder and light or a shadow are identified based on the correlation value of the change area 19 or the stored state of the edge information, thereby preventing erroneous determination.

[0011]

In the conventional image monitoring apparatus, as described above, the influence of light (removal of light) or the influence of shadow is removed (removal of light) depending on the correlation value of the change area or the stored state of the edge information. Even if an attempt is made to remove shadows, if a strong light such as sunlight enters the monitoring area, a false alarm is issued.

FIG. 3 shows that the light removal and the shadow removal are performed according to the stored state of the correlation value or the edge information of the change area.
It is a figure explaining the reason why when strong light like sunlight injects light or a shadow as an intruder, it is wrong. 3A shows the reference image 11, FIG. 3B shows the current image 16, and FIG.
8 is represented. When sunlight enters the monitoring area, the current image 16
A high-luminance light region 24 is generated above. Light area 2
Since the luminance of No. 4 is much higher than that of the other areas of normal luminance, it is impossible to detect the pattern on the floor 12 in the reference image 11. Therefore, the reference image 11 and the current image 16
Are low, and the edge information is not stored. Therefore, when the light region 24 moves,
It will be detected as an intruder.

When a shadow moves in the light area 24 due to a person passing outside the window or reflection of the movement of clouds,
The movement of the shadow is detected as an intruder, and an erroneous report is issued. The present invention provides an image monitoring apparatus that detects an intruder in a monitoring area by processing an image of the monitoring area to remove an erroneous detection due to light or shadow in the monitoring area and accurately detect the intruder. The purpose is to be able to.

[0014]

SUMMARY OF THE INVENTION The present invention has been made to achieve the above object. An image monitoring apparatus according to the present invention includes an imaging unit that captures an image of a monitoring area, and an image processing unit that detects an intruder in the monitoring area by processing the captured image. Different logic is used for the logic for extracting a high-luminance light region from the image and detecting an intruder in this light region and the intruder detection logic for other normal-luminance regions.

The image monitoring apparatus of the present invention detects an intruder by comparing a reference image with a current image and extracting a change area by an image processing means, as in the prior art. Here, when strong light such as sunlight is inserted into the monitoring area, the light area corresponding to the part where the light is inserted has much higher luminance in a captured image than a normal area.

The image monitoring apparatus of the present invention utilizes a difference in luminance in a photographed image to extract in advance a light region having a predetermined luminance or higher. Then, the intruder detection logic in the light area is made different from the intruder detection logic in the other normal luminance area. Thus, strong light such as sunlight entering from the window or a shadow caused by this light is distinguished from an intruder, and detection of the light or the shadow as the intruder is prevented. In addition, an intruder is reliably detected by a normal detection logic.

In the image monitoring apparatus of the present invention, the light removal processing is performed as the detection logic in the light area so as not to detect the light area as an intruder. The image processing means extracts a region where the luminance changes from a comparison between the captured images, and when the changed region occurs in the light region or includes the light region, the sensitivity for intruder detection is determined. Is lower than the sensitivity in the region of normal luminance. For example, when an intruder is detected when the change area moves by a set distance, the set movement distance in the light area is set longer than the set movement distance in the normal luminance area.

In the present invention, as the detection logic in the light area, shadow removal processing is performed so that the shadow in the light area is not erroneously detected as an intruder. The image processing unit extracts a region where the luminance changes from the comparison between the captured images, and the changed region is in the light region, has a lower luminance than the light region, and is in contact with the outer periphery of the light region. , The sensitivity for detecting an intruder is lower than the sensitivity in a region of normal luminance.

According to the present invention, in an image monitoring apparatus for detecting an intruder in a monitoring area by processing a captured image, erroneous detection due to light or shadow in the monitoring area is eliminated, and an intruder is accurately detected. Can be detected. If an intruder actually exists but is determined to be light or shadow, the security status will be normal and output to the outside even if there is an intrusion abnormality. Will occur. In order to prevent such a situation, it is possible to further limit the change of the detection logic in the light area. Also, if the light area is too large, the light removal and shadow removal logic can be disabled.

[0020]

Embodiments of the present invention will be described with reference to the drawings. FIG. 4 shows a configuration when the present invention is applied to an image sensor. The image sensor 31 is controlled by control means 32 including a CPU and the like. Imaging means 3
Reference numeral 3 includes an optical system and a CCD. When the monitoring area becomes dark, such as at night,
Under the control of the control means 32, an infrared ray is projected on the monitoring area. The image processing means 35 is configured by a CPU or the like,
The image acquired from the imaging unit 33 is processed to determine whether there is an intruder. This determination logic will be described later.
The contents of the storage means 36 will also be described later.

The operating means 37 is provided for setting or changing settings of the image sensor 31. The output unit 38 and the image output unit 39 are connected to a controller (not shown) by wire or wirelessly, and the controller is connected to a control center (not shown) via a communication line. When the image sensor 31 detects an intruder, an abnormal state is stored in the storage means 36 as a security state, and an abnormal image at the time of occurrence of an abnormality is stored. The output unit 38 and the image output unit 39 output the stored abnormal state and the abnormal image when the state of the state is called from the controller or the control center via the input unit (not shown). The power supply 40 converts AC power into DC power and supplies the DC power to each unit in the image sensor 31.

FIG. 5 shows the contents of the storage means 36. The storage means 36 has a program area 51 for storing operation programs of the control means 32 and the image processing means 35, a parameter area 52 for storing various parameters, and a reference image storage for storing the reference image photographed by the imaging means 33 and the current image. Area 5
3, the current image storage area 54, an abnormal image storage area 55 for storing an image when an intruder is detected, a work area 56 used for processing of the control means 32 and the image processing means 35, and a security state (normal, abnormal). A mode storage area 58 for storing a security mode (alert mode, release mode, maintenance patrol mode) of the image sensor, and a light area storage area 59 for storing a light area in a shooting screen are stored.

Next, the outline (principle) of the logic for detecting an intruder and the logic for light removal and shadow removal in the image processing means 35 will be described. The logic of intruder detection is
As described with reference to FIG. 1, it depends on the size of the change area 19 and the moving distance 23. The light removal logic will be described with reference to FIG.

6A is a reference image 11, FIG. 6B is a current image 16, FIG. 6C is a difference image 18 obtained by taking a difference between the reference image 11 and the current image 16, and FIG. Area 2
4 represents the extracted light area image 25. In the illustrated example, it is assumed that the light area 24 of the current image 16 is smaller and the brightness of the current image 16 is lower than that of the reference image 11. The image processing unit 35 extracts the light region 24 having a very high luminance from the luminance information of the reference image 11, and outputs the light region image 2 shown in (D).
5 is obtained. The light area image 25 is stored in the storage unit 3.
6 is stored in the optical area storage area 59. Next, the reference image 11 and the current image 16 are compared to obtain a luminance change area 19. If the change area 19 is generated in the light area 24 or includes the light area 24, the sensitivity for detecting an intruder is reduced. As an example of decreasing the sensitivity, it is determined that the intruder moves when the change area 19 moves by a set distance. However, when the change area 19 includes the light area 24 or is included in the light area 24, the set movement distance is set to the normal luminance area. Set longer than. This makes use of the fact that the change in luminance that appears due to the diurnal movement of sunlight, for example, is shorter than the intruder's movement distance within a certain period of time.

In the example shown in FIG. 6, since the area detected as the change area 19 is included in the light area 24, the light area 24 is not detected as an intruder if the above procedure is followed. If the light area 24 is very large, measures such as not operating the logic may be taken. The logic of shadow removal will be described with reference to FIG.

FIG. 7A shows the reference image 11, FIG. 7B shows the current image 16, FIG. 7C shows the difference image 18, and FIG.
Represents a light region image 25 obtained by extracting a light region 24 from the image data. In the illustrated example, a person passes outside the monitoring area, and the shadow 26 is simply projected into the light area 24, and there is no intruder in the monitoring area. In this example, when the reference image 11 and the current image 16 are compared, the shadow 26 appears in the difference image 18 and the shadow 2
6 moves.

As the shadow removal logic, the light region 24 is extracted from the luminance information of the reference image 11, and the light region image 25 shown in FIG. The light area image 25 is stored in the storage unit 3
6 is stored in the optical area storage area 59. Next, a change area 19 is obtained by comparing the reference image 11 and the current image 16.
The change region 19 is generated in the light region 24, has lower luminance than the light region 24, and has the light region 2
When touching the boundary of No. 4, the sensitivity for intruder detection is lowered. An example of lowering the sensitivity is the same as the above-described example of light removal. If the condition of ~ is satisfied,
The change area may be determined as a non-intruder.

In the example shown in FIG. 7, the changing area 19 satisfies the conditions of (1) to (5) for shadow removal, and the moving distance is short because the shadow 26 moves only within the light area 24. Therefore, the shadow 26 is not detected as an intruder. If the light area 24 is very large, measures such as not operating the logic may be taken. Next, the intruder detection processing of the image processing means 35 will be described in detail.

FIGS. 8 and 9 show intruder detection logic. This logic includes light removal and shadow removal logic. Before starting the intruder detection logic, the imaging means 33
The monitoring area is photographed in advance by using the reference image 1
1 is stored in the reference image storage area 53. Also,
From this reference image 11, a light area image 25 obtained by extracting the light area 24 is obtained in advance. This acquisition method will be described later with reference to FIG.

This logic is repeatedly executed, for example, at 0.5 second intervals. In step S 11, the current monitoring area is photographed by the imaging unit 33 to obtain the current image 16 and stored in the current image storage area 54. In step S12, a difference is obtained for each pixel between the reference image 11 and the current image 16, and binarized. When there is no predetermined luminance difference, the pixel value is set to 0, and when there is a predetermined luminance difference, the pixel value is set to 1. An area in which the pixels having the pixel value 1 are collected, that is, a set of the difference binarized areas is defined as a change area 19.
Becomes

In step S13, a moving object tracking process is performed.
This process tracks the same object on the image. For example,
When there are a plurality of change areas 19, the same change object 19 is tracked as the same object by associating the change area 19 of the past change area 19 and the current change area 19 which has moved a short distance. The new change area 19 that is not associated with the past change area 19 is a first-appearing moving object. If the change area 19 detected in the past is not tracked for, for example, 3 seconds, it is regarded as having disappeared, and the tracking processing is terminated.

In step S14, it is determined whether or not the change area 19 exists. If NO here, reference image 1
Since there is no change or slight change between 1 and the current image 16, the possibility of detecting an intruder is low, and the process proceeds to step S15 to end the process. If YES, there is a possibility that an intruder has been detected, and the process advances to step S21 to perform shadow removal processing.

In step S15, the current image 16 is stored in the reference image storage area 53 as the reference image 11. Thus, the reference image 11 is updated. In step S16, “normal state” is stored in the state storage area 57, and step S17
To update the light area 24 and end the processing. The updating of the light area 24 in step S17 will be described later in detail, but the light area 24 stored in the light area storage area 59 is updated to the light area of the current image 16 for each process. Go.

In step S21, since the change area 19 exists, it is determined whether or not the change area 19 satisfies the shadow condition. For the principle of shadow identification, see the description of FIG. 7 described above. If “NO” here, there is a possibility that an intruder has been detected, so the flow advances to step S31 to determine whether or not it is a light area detection. If “YES” in the step S21, the changing area 19 is likely to be a shadow and is unlikely to be an intruder, so the process proceeds to a step S22 of a process for preventing unreporting.

In step S31, since the change area 19 is not a shadow, it is determined whether or not this is the light area 24. For the principle of light identification, see the description of FIG. 6 described above. If "NO" here, there is a possibility that an intruder has been detected, so the flow advances to step 41 for intruder detection processing. If “YES” in the step S31, the changing area 19 is likely to be the light area 24 and is unlikely to be an intruder, so the process proceeds to a step S32 of a process for preventing a false alarm.

In step S41, to determine whether or not the change area 19 is an intruder, it is determined whether or not the change area 19 has moved 1.5 m or more from the first appearance position. That is, the first appearance position 20 in the reference image 11 and the current image 1
It is determined whether the moving distance 23 (see FIG. 1) between the change area 19 and the current position 22 in Step 6 is not less than 1.5 m.

When the presence of the change area 19 is detected in step S14, the reference image 11 is updated (step S1).
7) is not performed, so the first appearance position 20 of the change area 19
Remain stored in the reference image 11. Change area 19
If the intruder 17 has been detected, the current position 22 of the change area 19 moves with the movement of the intruder 17 by repeating the processing of this flowchart. The moving distance 23 between the first appearance position 20 and the current position 22 is 1.5 m
At this point, it is determined that the intruder 17 has been detected, and the process proceeds to step S42. If the moving distance 23 is less than 1.5 m, it is determined that the intruder 17 is still in a normal state, and the process proceeds to step S43.

In step S42, since the intruder 17 is detected, "intrusion abnormality" is stored in the state storage area 57, and the current image 16 is stored in the abnormal image storage area 55 as an abnormal image. On the other hand, in step S43, “normal state” is stored in the state storage area 57. Thereafter, in step S17, the light area 24 is updated, and the processing is terminated. Step S31
If YES in step S32, the change area 19 is likely to be the light area 24, but may be an intruder. Therefore, in step S32, the change area 19 is 3 m from the first appearance position 20.
It is determined whether or not it has moved. As described above, in the light removal processing, the threshold value of the moving distance 23 that determines the determination sensitivity is set to be longer than 1.5 m for intruder detection.
This makes it difficult for the light area 24 to be erroneously detected as the intruder 17. However, if the change area 19 is the intruder 17,
When moving from the first appearance position 20 by 3 m or more, it is detected as an intruder, and the process proceeds to step S33.

The processing in step S33 is similar to step S42, and the processing in step S34 is similar to step S43. Thereafter, in step S17, the light area 24 is updated, and the processing is terminated. If “YES” in the step S21, the changing area 19 is likely to be the shadow 26, but may be the intruder 17, so that in the step S22,
It is determined whether or not the change area 19 has moved from the first appearance position 20 by 4 m or more. As described above, in the shadow removal processing, the threshold value of the moving distance 23 that determines the determination sensitivity is set to be longer than 3 m for detecting the light area. This makes it difficult for the shadow 26 to be erroneously detected as the intruder 17. However, if the change area 19 is the intruder 17, it is detected as an intruder when moving from the first appearance position 20 by 4 m or more, and the process proceeds to step S23. Proceed to step S24.

The processing in step S23 is the same as step S42, and the processing in step S24 is the same as step S43. Thereafter, in step S17, the light area 24 is updated, and the processing is terminated. FIG. 10 shows the details of the processing for updating the light area 24 in step S17 in the intruder detection logic.

For updating the light area, a high-luminance pixel is detected for each pixel of the current image 16 and the light area image 25 is updated based on the result. In step S51, an absolute luminance (actual luminance in a monitoring area) is obtained from a luminance value and a camera exposure control value (aperture, shutter speed) for one pixel of the current image 16.

In step S52, the absolute luminance is 10,000.
It is determined whether it is greater than lux. The absolute luminance at a place where normal sunlight is not directly radiated is about several hundreds to 1000 lux, but the absolute luminance at a place where sunlight is directly radiated exceeds 10,000 lux. Therefore, the absolute luminance obtained from the luminance information of the pixels in the current image 16 is 1000
If it is larger than 0 lux, it is determined that the pixel is in the light area 24.

If the absolute luminance exceeds 10,000 lux,
In step S53, a value 32 is set to the value of the corresponding pixel of the light area image 25. Note that any value can be used for this numerical value. If the absolute luminance is equal to or less than 10000 lux, it is determined whether or not the value of the corresponding pixel of the light area image 25 is 0 in step S54. If the pixel value is 0, the process is terminated. If the pixel value is greater than 0, the process is terminated by subtracting 1 from the pixel value in step S55.

The above steps S51 to S55 are executed for all the pixels of the light area image 25. Thus, in the light region image 25, the value of the pixel included in the light region 24 becomes 1 or more, and the value of the pixel other than the light region 24 becomes 0. The reason why the value of the pixel included in the light region 24 in the light region image 25 is set to 32 is to prevent the light region 24 in the light region image 25 from being changed by a momentary change in luminance. Once it is determined that the pixel is in the light area 24, the pixel value is set to 32.
Until it is determined that the region is not the light region twice consecutively, the light region 24 is continued. That is, hysteresis is provided.

[0045]

According to the present invention, in an image monitoring apparatus for detecting an intruder in a monitoring area by processing an image of the monitoring area, erroneous detection due to light or shadow in the monitoring area is eliminated. , It is possible to accurately detect an intruder.

[Brief description of the drawings]

FIG. 1 is a diagram showing logic for detecting an intruder by an image monitoring device.

FIG. 2 is a diagram for explaining the reason for erroneously determining a light area as an intruder in the related art.

FIG. 3 is a diagram illustrating the reason for detecting light as an intruder according to the related art (part 1).

FIG. 4 is a diagram showing a configuration of an image sensor to which the present invention is applied.

FIG. 5 is a diagram showing contents of a storage unit in the image sensor of FIG. 5;

FIG. 6 is a view showing the principle of light removal in the image sensor of FIG. 5;

FIG. 7 is a view showing the principle of shadow removal in the image sensor of FIG. 5;

FIG. 8 is a flowchart (part 1) showing processing of the image processing means of FIG. 5;

FIG. 9 is a flowchart (part 2) showing the processing of the image processing means of FIG. 5;

FIG. 10 is a flowchart showing an operation of updating an optical area in the operations of FIGS.

[Explanation of symbols]

 11 ... Reference image 15 ... Window 16 ... Current image 17 ... Intruder 18 ... Difference image 19 ... Change area 20 ... First appearance position 22 ... Current position 23 ... Moving distance 24 ... Light area 25 ... Light area image 26 ... Shadow 31 ... Image sensor 32 ... Control means 33 ... Imaging means 34 ... Infrared light emitting means 35 ... Image processing means 36 ... Storage means 37 ... Operation means 38 ... Output means 39 ... Image output means 40 ... Power supply 51 ... Program area 52 ... Parameter area 53 ... Reference image storage area 54 ... Current image storage area 55 ... Abnormal image storage area 56 ... Work area 57 ... Status storage area 58 ... Mode storage area 59 ... Light area storage area

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Koichi Oza 8-10-16 Shimorenjaku, Mitaka City, Tokyo Inside Secom Co., Ltd. (72) Inventor Tetsuya Takahashi 6-11-23 Shimorenjaku, Mitaka City, Tokyo Seco (72) Inventor Yoshiki Kobayashi 7-1-1, Omika-cho, Hitachi City, Ibaraki Prefecture Inside Hitachi Research Laboratory, Hitachi, Ltd. (72) Inventor Tomoko Onuma 7-1-1, Omika-cho, Hitachi City, Ibaraki F-term in Hitachi Research Laboratory, Hitachi, Ltd. F-term (reference)

Claims (4)

[Claims] 1. An image monitoring apparatus comprising: an image capturing unit that captures an image of a monitoring area; and an image processing unit that detects an intruder in the monitoring area by processing the captured image. Different logic is used for extracting a high-luminance area from an image, and using logic for detecting an intruder in the high-luminance area and logic for detecting an intruder in other normal-luminance areas. Image monitoring device. 2. The image processing means extracts a change in luminance from a comparison between captured images, and when the luminance change occurs in the high luminance area, or when the luminance change includes the high luminance area. The image monitoring apparatus according to claim 1, wherein in the case, the sensitivity for detecting the intruder is lower than the sensitivity in the normal luminance region. 3. The image processing means extracts a change in luminance from a comparison between captured images, and the luminance change is within the high-luminance area, and is lower than the luminance of the high-luminance area. The image monitoring apparatus according to claim 1, wherein when in contact with an outer periphery of the high luminance area, sensitivity for detecting the intruder is lower than sensitivity in the normal luminance area. 4. The image processing means extracts a luminance change from a comparison between captured images, wherein the luminance change is within the high luminance area, has a lower luminance than the luminance of the high luminance area, and The image monitoring device according to claim 1, wherein the image monitoring device is a non-intruder when in contact with an outer periphery of the luminance area.