JP2003134504A - Mask detector and monitor camera system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a mask discrimination device that detects it when a monitor camera system is masked and to provide the monitor camera system provided with the mask discrimination device. SOLUTION: The mask discrimination device discriminates that the monitor camera system is masked (S5) when an area to be monitored has a certain degree of illuminance or more (S1); the image of monitor is a flat image because the number of edges extracted from the image is a few (S2); the number of edges of the present image is greatly smaller than that of the original image (S3); and the gain to adjust the level of an image signal is a reference value or more (S4). This invention discriminate the presence/absence of the mask by comparing the aperture of an imaging section and a shutter speed or the like with a reference value or a threshold value in addition to the illuminance, the number of the edges, the increase/decrease in the edges, and the gain. When the mask discrimination device discriminates that the monitor camera system is masked, the monitor camera system alerts the occurrence of the masked monitor camera system.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mask detection device for detecting that a surveillance camera device for photographing a surveillance region has a mask for obstructing photographing, and a surveillance camera device using the mask detection device. .

[0002]

2. Description of the Related Art Conventionally, a surveillance camera is installed in a store or a doorway which is a surveillance place, an image transmitted from this surveillance camera is displayed on a monitor at a surveillance center, and this image is recorded to be used as ex post evidence. Surveillance camera devices are known.

[0003]

In the above-mentioned conventional surveillance camera device, an illegal person attaches a piece of tape or the like to the surveillance camera to mask it, thereby obstructing the field of view of the surveillance camera and taking an image of a criminal act. Sometimes you can't. Such fraudulent activity can be noticed if the surveillance staff constantly monitors the monitor, but if the surveillance camera is only recording the image, the fraudulent activity will not be noticed and the image of the criminal behavior will be displayed. I couldn't record.

It is an object of the present invention to provide a mask detection device capable of detecting a mask on a surveillance camera device. Moreover, it aims at providing the surveillance camera apparatus provided with this mask detection apparatus.

[0005]

The present invention has been made to achieve the above object. The first mask detection device of the present invention is applied to the surveillance camera device when the edges extracted from the current image of the surveillance region are sharply reduced from the original image and the gain for image signal level adjustment is smaller than the reference value. It is determined that the mask is applied. This is to determine that the surveillance camera device has been masked when the image suddenly becomes flat and dark. In this case, false alarm can be eliminated by setting the condition that the number of edges is smaller than the threshold value and that the illuminance of the monitoring area is to some extent.

In the second mask detecting apparatus of the present invention, the difference between the first reference gain value and the current gain value is equal to or more than a predetermined threshold value, and the current gain value and the second reference gain value are equal to each other. When the absolute value of the difference is greater than or equal to a predetermined threshold value, it is determined that the surveillance camera device is masked. This is to determine that the mask has been applied when the image suddenly darkens. Also in this case, false alarm can be eliminated by setting the condition that the number of edges is less than the threshold value and that the illuminance of the monitoring area is to some extent.

In the first and second mask detecting devices,
Various reference values are used. The values of these criteria are
The image captured by the imaging unit is updated when the image changes normally. As a result, a difference is extracted from the masked abnormal image and the reference value, and accurate mask determination can be performed.

According to the third mask detecting apparatus of the present invention, when the illuminance of the current image is equal to or higher than a predetermined threshold value and the current gain value is equal to or higher than the predetermined threshold value, the mask is set on the surveillance camera apparatus. It is determined that it has been done. This is to determine that the mask has been applied in an unmatched state in which the captured image is dark although the ambient light being captured is sufficiently bright. Also in this case, false alarm can be eliminated by setting the condition that the number of edges of the monitoring area is smaller than the threshold value.

According to a fourth mask detecting apparatus of the present invention, when the illuminance of the monitoring area is less than a predetermined threshold value and the aperture value of the image pickup section is more than the predetermined threshold value, the mask is set on the monitoring camera apparatus. It is determined that it has been done. This is to determine that the illuminance sensor has been masked when it is detected that the captured image is sufficiently bright but the illuminance sensor is dark.

In the present invention, as the fifth mask detecting device, there is a case where the mask determination is performed by using the projection and reception of infrared rays. This device determines that the mask is made by projecting infrared rays and detecting the infrared rays reflected by the mask.

In a sixth mask detecting apparatus of the present invention, the current gain value is the maximum value, the average luminance of the entire current image is less than a predetermined threshold value, the image is divided into a plurality of blocks, When the number of blocks whose average brightness is less than the predetermined threshold value is equal to or more than the predetermined threshold value, it is determined that the mask is performed. This is to judge that the mask is applied when it is detected that the current image is dark although the monitoring area is bright. Also in this case, false alarm can be eliminated by setting a condition that the illuminance of the monitoring area is to some extent.

A seventh mask detecting apparatus of the present invention masks the surveillance camera apparatus when the illuminance of the monitoring area is equal to or higher than a predetermined threshold value and the shutter speed value of the image pickup section is equal to or higher than the predetermined threshold value. It is determined that the This is to judge that the mask is applied when it is detected that the current image is dark although the monitoring area is bright.

In the eighth mask detecting apparatus of the present invention, the illuminance of the monitoring area is equal to or higher than a predetermined threshold value, the current gain is 0, and the number of edges of the current image is less than the predetermined threshold value. Yes, when the difference between the maximum value and the minimum value of the average luminance of a plurality of blocks of the current image is less than a predetermined threshold value, it is determined that the surveillance camera device is masked. This is to judge that the mask is applied when it is detected that the monitoring area is bright and the current image is flat.

The surveillance camera apparatus according to the present invention includes the above-mentioned first to third aspects.
At least one of the mask detection devices 4 and 6 to 8 is provided. Further, other devices such as the fifth mask detection device described above can be combined. The surveillance camera apparatus of the present invention issues a mask abnormality when any one of the mask detection apparatuses detects a mask.

[0015]

DETAILED DESCRIPTION OF THE INVENTION An embodiment of the present invention will be described with reference to the drawings. The overall configuration of a monitoring system adopting the present invention will be described with reference to FIG. A plurality of surveillance camera devices 1 are arranged in the surveillance area to take an image of the surveillance place. The surveillance camera device 1 is connected to the controller 2. A monitor 3 and a speaker 4 are attached to the controller 2.

Communication between the surveillance camera device 1 and the controller 2 such as setting data and camera connection state is carried out.
The image captured by the surveillance camera device 1 is transmitted to the controller 2 at the same time as the above communication. The controller 2 records and reproduces images by using a hard disk. When an abnormality occurs in the monitoring area, the monitoring camera device 1 notifies the controller 2 of the abnormality and transmits an image when the abnormality is detected by image processing. The controller 2 issues an alarm sound from the speaker 4 and displays an abnormality warning on the monitor 3 to notify the outside of the abnormality.

Further, the surveillance camera device 1 holds several images in advance inside the device. Note that among the held images, the latest image is called “current image” and the images older than that are called “original image”. If an abnormality occurs while the controller 2 is not in the state of recording an image, the image held in advance is sent to the controller 2. As a result, the controller 2 can confirm the abnormality by comparing the current image before the occurrence of the abnormality, the time when the abnormality occurred, and the current image,
Evidence images can be recorded.

The configuration of the surveillance camera device 1 will be described with reference to FIG. The lens 5 and the image pickup device 6 form an image pickup unit 7. The image sensor 6 is composed of a CCD. The image formed on the image sensor 6 through the lens 5 is converted into an electric signal and input to the control unit 8. The image signal can be output to the outside through the monitor I / F (interface) 9. The communication unit 11 communicates with the controller 2. When an abnormality occurs, the controller 2 displays an abnormality.

The storage unit 12 stores a part of the image signal data and also stores a plurality of images in a ring buffer form. An illuminance sensor 13 and an infrared light receiving sensor 15 are provided to input external environment data to the control unit 8. An infrared projector 14 for mask detection is provided. Display unit 16
Is configured by an LED or the like, and displays the state of the surveillance camera device 1 to the outside.

The control section 8 has an image pickup control means 17, a display / output means 18, a mask determination means 19, and an image processing means 21. The image pickup control unit 17 controls the aperture of the lens 5 of the image pickup unit 7, controls the shutter speed of the image pickup device 6, and controls the gain so that the image signal level becomes constant. The aperture value is set to a large value when the monitored area becomes bright. The shutter speed value is set to a small value when the monitored area becomes bright. The gain has a small value when the monitored area becomes bright. Further, when the illuminance of the monitoring area is bright, the control is performed by the diaphragm, when the illuminance is dark, the control is performed by the shutter speed, and when the illuminance is intermediate, the control is performed by the gain.

The display / output means 18 performs processing for outputting an image and other information from the display section 16, the monitor I / F 9, and the communication section 11. The image processing unit 21 determines whether or not there is an abnormality by performing image processing, and when detecting the abnormality, outputs an abnormality signal from the communication unit 11 to the controller 2. Further, the average brightness of the entire image, the average brightness of each image when the image is divided into six blocks, and the number of edges in the image are extracted.

The mask determination means 19 includes an illuminance sensor 13
Detected, the infrared level received by the infrared light receiving sensor 15, the average brightness of the entire image extracted by the image processing means 21, the average brightness of each of the blocks obtained by dividing the image into a plurality of blocks, and the number of edges, and imaging. A gain value for adjusting the image signal level output from the unit 7 to a predetermined level, an aperture value output to the imaging unit 7, and a shutter speed are input. The mask determination means 19 comprehensively determines whether or not the surveillance camera device 1 is masked based on these pieces of information. In addition, the infrared projector 14 controls the projection of infrared rays.

(Mask judgment process) Referring to FIGS.
First to eighth determination processing in the mask determination means 19 will be described. The following processing is executed every predetermined time. In each of the following determination processes, the M1 to M8 flags are turned on when it is determined that the mask is applied, and the M1 to M8 flags are turned off when the mask is not determined to be applied. This M
The 1 to M8 flags are used in the alarm processing described later.

(Mask Judgment Process 1) Referring to FIG.
The mask determination processing of will be described. Here, it is determined that the mask is pasted mainly because the number of edges in the image is reduced. If the image is a photograph of the surveillance area, many edges are extracted because various things are reflected in the image. On the other hand, when a mask is attached in front of the lens 5, the image of the mask becomes flat and the number of edges is reduced.

In step S1, the illuminance (S illuminance) of the current monitoring area detected by the illuminance sensor is the illuminance threshold (CT
It is determined whether or not h1) or more. The relationship between CTh1 and other illuminance threshold values is shown in FIG. If a bright image is not obtained to the extent that mask determination is possible, and the environment is such that the following mask determination processing can be performed, it becomes NO,
In step S6, the M1 flag is turned off. If YES, the process proceeds to step S2.

In step S2, it is determined whether or not the number of edges (S edges) extracted from the current image by the image processing means 21 is less than the edge threshold value (ETh1). E
The relationship between Th1 and other edge thresholds is shown in FIG. Here, it is determined whether or not the image is flat. When the mask is placed in front of the lens 5 as described above, the image becomes flat. If it is not flat, the result is NO, the process proceeds to step S6 to turn off the M1 flag, and if it is flat, the result is YES and the process proceeds to step S3.

In step S3, the number of edges (S edges) obtained from the current image is the reference number of edges (R) obtained from the original image.
It is determined whether it is smaller than the edge) by -ETh2 (%) or more. The relationship between ETh2 and other edge thresholds is shown in FIG.
6 is shown. If the number of edges of the current image has sharply decreased from the original image, the determination result is YES and the process proceeds to step S4. NO
If so, the process proceeds to step S6 to turn off the M1 flag.

In step S4, it is determined whether or not the gain (S gain) of the image pickup control means 17 is equal to or larger than the reference gain (R gain) obtained from the original image. The result is NO if the current image is brighter than the original image. Since the image does not become bright when masked, in this case, the process proceeds to step S6 and the M1 flag is turned off. If YES, the image is dark, so the flow advances to step S5 to turn on the M1 flag. The turning on of the M1 flag means that the masking has been performed.

According to the processing of FIG. 3 described above, the phenomenon in which the image is flat due to the masking of the lens 5 is used, and when the number of edges extracted from the image is small and the edge sharply decreases. , It is determined that the mask is applied. In addition, false alarms are prevented by setting the condition that the illuminance of the monitoring area is a little and that the image is not bright.

(Mask determination process 2) The second process will be described with reference to FIG.
The mask determination processing of will be described. Here, the fact that the image is abruptly darkened when the mask is applied is used to detect that the mask is attached. In step S7, the illuminance (S illuminance) detected by the illuminance sensor is the illuminance threshold (CT
It is determined whether or not h1) or more. The relationship between CTh1 and other illuminance threshold values is shown in FIG. N if the image is not bright enough to determine the mask
When it becomes O, the process proceeds to step S12 and the M2 flag is turned off. If YES, the process proceeds to step S8.

In step S8, the number of edges of the current image (S
Edge) is less than the edge threshold (ETh3). The relationship between ETh3 and other edge thresholds is shown in FIG. If the number of edges is equal to or less than the normal value (is flat), the determination result is YES and the process proceeds to step S9. If NO, the M2 flag is turned off in step S12. In step S9, it is determined whether or not the difference between the reference gain (R gain I) determined by the original image and the gain (S gain) of the current image is equal to or larger than the gain threshold value (GTh1). The relationship between GTh1 and other gain thresholds is shown in FIG. If the image suddenly becomes darker than the original image, YES is determined and the step S10 is performed.
Go to. If NO, the M2 flag is turned off in step S12,

In step S10, the gain value (S
It is determined whether the absolute value of the difference between the gain) and the normal gain when dark (R gain II) is greater than or equal to the gain threshold (GTh2). The relationship between GTh2 and other gain thresholds is shown in FIG. S gain and R gain I
If I is almost the same, NO is determined, and M is determined in step S12.
Turn off the 2 flag to prevent false alarms. If yes,
In step S11, the M2 flag is turned on.

According to the processing of FIG. 4 explained above, when the lens 5 is masked and the image suddenly becomes dark, it is determined that the masking is performed. In addition, false alarms are prevented under the conditions that the illuminance of the monitoring area is a little, that there is an edge of the image, and that the gain is almost the same as in the dark environment.

(Mask determination process 3) The third process will be described with reference to FIG.
The mask determination processing of will be described. Here, it is determined that the mask is attached when the current image is dark although the ambient light in the monitoring area is bright. In step S13, it is determined whether the illuminance detected by the illuminance sensor (S illuminance) is equal to or higher than the illuminance threshold (CTh2). The relationship between CTh2 and other illuminance thresholds is shown in FIG. If a sufficiently bright image has not been obtained, the determination is NO, and the process advances to step S17 to turn off the M3 flag. If YES, step S
Proceed to 14.

In step S14, it is determined whether or not the number of edges (S edges) of the current image is less than the edge threshold value (ETh4). The relationship between ETh4 and other edge thresholds is shown in FIG. If the current image is flat to some extent, the determination is YES and the process proceeds to step S15. If NO, the process advances to step S17 to turn off the M3 flag.

In step S15, it is determined whether or not the gain (S gain) of the current image is equal to or larger than the gain threshold value (GTh3). The relationship between GTh3 and other gain thresholds is shown in FIG. If the taken image is dark, the determination result is YES, and the process proceeds to step S16 to turn on the M3 flag. If NO, the process advances to step S17 to turn off the M3 flag. According to the processing of FIG. 5 described above, when the unmatched occurrence that the captured image is dark even though the ambient light is sufficiently bright, it is determined that the mask is applied.
In addition, by setting a condition that the current image is a flat image, it is possible to prevent an erroneous report when there is a possibility that some image is captured without being masked.

(Mask determination process 4)
The mask determination processing of will be described. Here, it is determined that the mask is attached when the captured image is sufficiently bright but the illuminance sensor does not detect sufficient illuminance. In step S18, it is determined whether the illuminance (S illuminance) detected by the illuminance sensor 13 is less than the illuminance threshold (CTh3). The relationship between CTh3 and other illuminance threshold values is shown in FIG. If the dark illuminance is not obtained, the determination is NO and the process proceeds to step S21 to turn off the M4 flag. If YES, the process proceeds to step S19.

In step S19, it is determined whether or not the diaphragm (S diaphragm) at the time of photographing the current image is equal to or larger than the threshold value ITh1. If the current image is dark, the result is NO, and in step S21 M
4 Turn off the flag. If YES, step S20
Turns on the M4 flag. According to the processing of FIG. 6 described above, it is determined that the mask is applied when an unmatch occurs that the image is bright but the illuminance detected by the illuminance sensor 13 is dark. In the above-described first to third mask determination processing, mask detection is not performed based on the detection result of the illuminance sensor. Therefore, it becomes important to detect the masking action of the illuminance sensor 13 as in this example.

(Mask Determination Processing 5) The fifth mask determination processing will be described. Conventionally, a method has been proposed in which infrared rays are projected by the infrared projector 14 in front of the surveillance camera apparatus 1 and reflected infrared rays are received by an infrared light receiving sensor 15 to determine whether or not a mask is applied (example). , Japanese Patent Application No. 2001-275182). This device is also used in the surveillance camera device 1 of this example, and when the mask is attached, the M5 flag is turned on.

(Mask Judgment Process 6) The sixth process will be described with reference to FIG.
The mask determination processing of will be described. Here, it is determined that the mask is attached when the captured image is dark although the environment is bright. In step S22, it is determined whether or not the illuminance (S illuminance) detected by the illuminance sensor 13 is equal to or higher than the illuminance threshold (CTh1). The relationship between CTh1 and other illuminance threshold values is shown in FIG. If a bright image to the extent that mask determination is possible has not been obtained, the determination is NO, and the flow advances to step S27 to turn off the M6 flag. YE
If S, go to step S23.

In step S23, it is determined whether or not the gain (S gain) of the current image has the maximum value. If the taken image is dark, the determination result is YES and the process proceeds to step S24.
If NO, the M6 flag is turned off in step S27. In step S24, it is determined whether the average brightness of the entire image extracted by the image processing means 21 is less than the brightness threshold YTh. The relationship between YTh1 and other luminance threshold values is shown in FIG. If the taken image is dark, the determination result is YES, and the process proceeds to step S25. If NO, step S
At 27, the M6 flag is turned off.

In step S25, it is determined whether or not the average brightness of each of the blocks obtained by dividing the image extracted by the image processing means 21 into a plurality of values is less than the brightness threshold value YTh2. Then, it is determined whether or not the number of blocks that is less than or equal to is the number threshold value NTh1 or more. The relationship between YTh2 and other luminance threshold values is shown in FIG. If the taken image is dark, the determination is YES, and the M6 flag is turned on in step S26. If NO, the M6 flag is turned off in step S27.

According to the processing of FIG. 7 described above, it is determined that the mask is applied by utilizing the phenomenon that the photographed image is black although the environment is bright. This process is black
This is effective when detecting that a mask is used with something that does not transmit light. In step S25, the image is divided into several blocks, and when the black blocks occupy most of the blocks, it is determined that the mask is abnormal. This improves the detection accuracy.

(Mask determination process 7) The seventh process will be described with reference to FIG.
The mask determination processing of will be described. Here, it is determined that the mask is attached when the environment is bright and the created image is dark. In step S28, it is determined whether or not the illuminance (S illuminance) detected by the illuminance sensor is equal to or higher than the illuminance threshold (CTh1). The relationship between CTh1 and other illuminance threshold values is shown in FIG. If a bright image to the extent that mask determination is possible has not been obtained, the determination is NO, and the flow advances to step S31 to turn off the M7 flag. If YES, the process proceeds to step S29.

In step S29, it is determined whether or not the shutter speed value (S shutter speed value) at the time of shooting the current image is less than the threshold value STh1. As described above, the slow shutter (large shutter speed value) is used when the image is dark even if the aperture is opened and the gain is increased. If the current image is dark, the determination is YES, and the M7 flag is turned on in step S30. If NO, the M7 flag is turned off in step S31. According to the processing of FIG. 8 described above, it is determined that the mask is attached when the environment is bright and the image is dark.

(Mask determination process 8) The eighth process will be described with reference to FIG.
The mask determination processing of will be described. Here, it is determined that the mask is attached when the image captured in a bright environment is flat. In step S32, the illuminance (S illuminance) detected by the illuminance sensor 13 is the illuminance threshold (C
Th4) or more is determined. The relationship between CTh4 and other illuminance threshold values is shown in FIG.
If a sufficiently bright image has not been obtained, the result is NO, the process proceeds to step S38, and the M8 flag is turned off. If YES, the process proceeds to step S33.

In step S33, it is determined whether or not the infrared level received by the infrared light receiving sensor 15 when the infrared light projector 14 is not projecting light is equal to or higher than the threshold IRTh1. If infrared rays in the ambient light are detected, the result is YES, it is determined that the environment is bright, and the process proceeds to step S34. If NO, the process advances to step S38 to turn off the M8 flag.

In step S34, it is determined whether or not the gain (S gain) of the current image is 0 or less. If the image is bright, the determination is YES and the process proceeds to step S35. If NO, the process advances to step S38 to turn off the M8 flag. In step S35, it is determined whether or not the number of edges (S edges) obtained from the current image is less than the threshold value ETh5. The relationship between ETh5 and other edge thresholds is shown in FIG. Y if the shot image is very flat
It becomes ES, and the process proceeds to step S36. If NO, the process advances to step S38 to turn off the M8 flag.

In step S36, the image processing means 21 obtains the luminance value of each block obtained by dividing the current image into six blocks, and the difference between the maximum value and the minimum value of each luminance value (MAX-MI).
It is determined whether N) is less than the brightness threshold value YTh3. The relationship between YTh3 and other luminance threshold values is shown in FIG. If the image is flat, the determination is YES and the process proceeds to step S37 to turn on the M8 flag. N
If it is O, the process proceeds to step S38 to turn off the M8 flag.

According to the processing of FIG. 9 described above, when the environment is sufficiently bright and the image is flat, it is determined that the mask is attached. This processing is effective when it is detected that the mask is made of a material having a high light transmittance or a white material.

(Issuing Report) As described with reference to FIGS. 3 to 9, the mask determination means 19 turns on at least one of the M1 to M8 flags when it determines that the mask has been made. In response to this, the display / output means 18 issues a mask abnormality to the outside. At this time, if the M1 to M8 flags are turned on to immediately issue an alarm, an abnormal mask is issued only by a momentary image change without a mask action (erroneous alarm).
There is a risk. On the other hand, if there is a true masking action, the masking action continues for at least a few seconds. In this example, the mask abnormality is output only when the M1 to M8 flags continue for several seconds.

FIG. 10 shows the alert determination processing in the display / output means 18. In step S39, it is determined whether or not any of the flags M1 to M8 is turned on. If NO here, the timer TM1 is reset (returned to the initial value) and the MASK flag is turned off in step S41. If YES, the process proceeds to step S40.

The timer TM1 is decremented in step S40, and the value of the timer TM1 is set to 0 in step S42.
Is determined. The timer TM1 is shown in FIG.
The initial value is set so that the value becomes 0 when the process of is repeated for 10 seconds. If NO, step S44
Then, the MASK flag is turned off. If the timer TM1 has timed out, the result is YES, and the process proceeds to step S43.
Turn on the ASK flag. When the MASK flag is turned on, a mask abnormality is reported.

In the example shown in FIG. 10, mask determination processing 1
Although the alerting process is performed using all of 8 to 8, mask determining processes other than the above may be added depending on the characteristics of the monitoring target area, or a part of the mask determining processes 1 to 8 may be omitted. You can also For example, when the monitoring target is a dark place such as a warehouse, mask determination processing 1 (detecting that the edge of the monitoring image has suddenly decreased), 5 (detection of infrared reflection), 6
Alternatively, a combination of 7 (detecting that the surveillance image has become completely black) is preferable.

When it is a place such as a pachinko parlor that is bright during business and has some surveillance camera illumination turned on at night, mask determination processing 1 (detection that the surveillance image edge has suddenly decreased), 2 A combination of (detecting that the surveillance image has become abruptly dark) and 6 or 7 (detecting that the surveillance image has become completely black) are preferable. If it is always a bright place such as a 24-hour ATM or convenience store, mask determination processing 1 (detection that the monitor image edge has rapidly decreased), 2
(Detects that the surveillance image has suddenly become dark), 3 (the image is dark and the illuminance sensor is bright and detects an unmatched state)
Is preferred. In this case, the mask determination process 4 (detecting the mask of only the illuminance sensor) can be employed as an auxiliary role.

If it is very bright in the daytime, as in outdoor monitoring, the mask determination processing 1 (detects that the edge of the monitoring image has suddenly decreased), the same 2 (detects that the monitoring image has suddenly become dark), A combination of 8 (detecting that the image has no shading) is preferable. In this case, the mask determination process 4 (detecting the body mask of the illuminance sensor) can be further employed as an auxiliary role.

(Update of Reference Value and Initial Setting) In mask determination processing 1 and 2, R illuminance, R edge,
Various reference values such as R gain are used. Since these reference values are acquired from the image (original image) at the time point before the current image is captured or the detection value of the illuminance sensor 13, they always fluctuate. For this reason,
When the mask determination process is started, the reference value must be updated. However, if there is a possibility of masking, the reference value should not be updated even if there is a change.

(Initial acquisition of reference value) As shown in FIG. 11, when the power of the surveillance camera device 1 is turned on,
Initial reference value acquisition is performed. Image sensor 6
CCD is not stable immediately after the power is turned on. Therefore, the reference value is fetched after a lapse of a predetermined time after the power supply to the CCD is stabilized.

When the power is turned on (step S45),
In step S46, the timer TM2 is reset (returned to the initial value). The initial value of the timer TM2 is step S4.
The value is set to 0 when 7 and 48 are repeated for 10 seconds. In step S48, it is determined whether or not the value of the timer TM2 is 0, and if not 0, step S4
Decrement the value of timer TM2 in step 7,
The process of returning to 8 is repeated.

When 10 seconds have elapsed and the value of the timer TM2 becomes 0, the process proceeds to step S49, and each reference value data is fetched from the photographed image, the illuminance sensor and the like. Then, each determination process is started. According to the processing of FIG. 11, the reference value acquisition processing is performed when the CCD becomes stable after 10 seconds have passed since the power was turned on.

(Update of Reference Value) FIGS. 12 to 14
Indicates a process of updating the reference value. This process is executed every predetermined time. In step S50, it is determined whether the absolute value of the difference between the reference illuminance (R illuminance) and the illuminance (S illuminance) of the current monitoring area is less than a predetermined threshold value RCTh1. The relationship between RCTh1 and other illuminance threshold values is shown in FIG. YE when the illuminance change is small
The result is S, and the process proceeds to step S54. If NO, the process proceeds to step S51.

The timer TM3 is decremented in step S51, and it is determined in step S52 whether or not the value of the timer TM3 has become zero. If NO here, the process proceeds to another process (step S64). If the illumination change has continued, YES is determined and the process proceeds to step S53. In addition, TM3
And the relationship with other timer values are shown in FIG. In step S53, the reference edge (R edge), the reference gain (R gain I, II), and the reference illuminance (R illuminance) are updated. That is, the current edge, gain, and illuminance are taken in as reference values. According to steps S51 to 53, each reference value is updated when the illuminance continues to change.

If YES in step S50 (small change in illuminance), the reference illuminance (R illuminance) is updated in step S54. That is, the current value is fetched. Next, in step S55, the timer TM3 is reset (returns to the initial value), and the process proceeds to step S56. Step S56
Then, it is determined whether or not the difference between the reference gain (R gain I) and the current gain (S gain) is greater than or equal to the threshold value RGITh1. The relationship between RGITh1 and other gain thresholds is shown in FIG. If the image becomes brighter than a certain value, YES is obtained and the process proceeds to step S61.
If NO, the process proceeds to step S57.

The timer TM4 is decremented in step S61, and it is determined in step S62 whether the timer TM4 has become 0 or not. The relationship between TM4 and other timer values is shown in FIG. If the image continues to be bright, the determination result is YES and the process proceeds to step S63.
If NO, the process proceeds to step S64. Step S63
Then, the reference edge (R edge) and the reference gain (R gain I, II) are updated. After that, it advances to step S64.

If NO at step S56 (when the image becomes dark or slightly bright), step S5
At 7, the timer TM4 is reset. Step S58
Then, it is determined whether or not the difference between the reference gain (R gain I) and the current gain (S gain) is greater than or equal to the threshold value RGITh2. The relationship between RGITh2 and other gain thresholds is shown in FIG. If the image becomes slightly brighter, YES is obtained and the reference gain (R gain II) is updated in step S59. In this update,
The current reference gain (R gain I) is captured in R gain II.

NO in step S58 and step S59
After that, in step S60, the reference gain (R gain I
It is determined whether or not the absolute value of the difference between I) and the current gain (S gain) is less than the threshold value RGIITh1. R
The relationship between GIITh1 and other gain thresholds is shown in FIG. If YES here, the updating process in step S63 is performed, and if NO, the process proceeds to step S64.

In step S64, the difference between the current edge number (S edge) and the reference edge (R edge) is the threshold value RE.
It is determined whether or not it is less than Th1. As a result, it is determined whether the increase amount of the edge is small. Step S65
Then, it is determined whether or not the value obtained by subtracting the ratio of the S edge and the R edge from 1 is less than the threshold value RETh2 (%).
As a result, it is determined whether or not the edge reduction amount is small.

If YES in both steps S64 and 65, the reference edge (R edge) is updated in step S66, and the process proceeds to step S67. If either is NO,
Step S66 is skipped and the process proceeds to step S67.
In step S67, the absolute value of the difference between the reference gain (R gain I) and the current gain (S gain) is the threshold value RGITh.
It is determined whether it is less than 3. The relationship between RGITh3 and other gain thresholds is shown in FIG. If the change in gain is small, the determination is YES, the reference gain (R gain I) is updated in step S68, and the process ends. N
If it is O, the process is terminated without updating.

According to the update process described above, each reference value is updated when it is determined that the illuminance of the monitoring area, the original image, and the like have changed normally. In addition, each reference value is not updated when an abnormal change occurs. This allows
Whether or not the mask is attached can be processed using a normal reference value.

FIGS. 15 to 20 show the relationship between the threshold values. FIG. 15 shows the relationship between the threshold values of the illuminance sensor 13. CTh1 to CTh have different brightness. Since RCTh1 is the amount of change with respect to the reference value, it is different from the others. FIG. 16 shows the relationship between the edge thresholds. ETh1, 3 to 5 have different degrees of flatness. ETh2 is a threshold value that determines how much the reference value has decreased, and is different from the others.

FIG. 17 shows the relationship of gain threshold values for mask determination. GTh1 and 2 have different sizes. GTh3 is unlike any other. FIG.
Shows the relationship of the gain threshold for updating the reference value. RGITh1, RGITh2 and RGITh1 are different in size.

FIG. 19 shows the relationship between the brightness threshold values. There is a difference in darkness between YTh1 and YTh2. YTh3 is unlike any other. FIG. 20 shows the relationship between timers. There is a difference in length between TM1 and 2 and TM3 and 4.

[0073]

According to the present invention, it is possible to provide a mask determination device capable of detecting a mask on the surveillance camera device. Further, it is possible to provide a surveillance camera device provided with this mask determination device.

[Brief description of drawings]

FIG. 1 is a diagram showing an overall configuration of a monitoring system adopting the present invention.

FIG. 2 is a configuration diagram of the surveillance camera device of FIG.

FIG. 3 is a flowchart showing mask determination processing 1 of the mask determination means in FIG.

FIG. 4 is a flowchart showing mask determination processing 2 of the mask determination means in FIG.

5 is a flowchart showing mask determination processing 3 of the mask determination means in FIG.

FIG. 6 is a flowchart showing mask determination processing 4 of the mask determination means in FIG.

7 is a flowchart showing mask determination processing 6 of the mask determination means in FIG.

8 is a flowchart showing mask determination processing 7 of the mask determination means in FIG.

9 is a flowchart showing a mask determination process 8 of the mask determination means in FIG.

FIG. 10 is a flowchart showing a reporting process of the display / reporting means of FIG.

11 is a flowchart showing an initial reference value fetching process in the mask determination means of FIG.

FIG. 12 is a flowchart (No. 1) showing the process of updating the reference value in the mask determination means of FIG.

FIG. 13 is a flowchart (No. 2) showing the process of updating the reference value in the mask determination means of FIG.

FIG. 14 is a flowchart (No. 3) showing the process of updating the reference value in the mask determination means of FIG.

FIG. 15 is a diagram showing a relationship between illuminance sensor threshold values in the mask determination means of FIG.

16 is a diagram showing a relationship between edge thresholds in the mask determination means of FIG.

17 is a diagram showing a relationship of gain threshold values in the mask determination means of FIG.

FIG. 18 is a diagram showing a relationship between gain threshold values in the mask determination means of FIG.

FIG. 19 is a diagram showing a relationship between luminance threshold values in the mask determination means of FIG.

20 is a diagram showing a relationship between timer values in the mask determination means of FIG.

[Explanation of symbols]

1 ... Surveillance camera device 2 ... Controller 3 ... Monitor 4 ... speaker 5 ... Lens 6 ... Image sensor 7 ... Imaging unit 8 ... Control unit 9 ... Monitor interface 11 ... communication section 12 ... Storage unit 13 ... Illuminance sensor 14 ... Infrared projector 15 ... Infrared light receiving sensor 16 ... Display 17 ... Imaging control means 18 ... Display / output means 19: Mask determination means 21 ... Image processing means

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G08B 25/04 G08B 25/04 E H04N 5/225 H04N 5/225 C F term (reference) 5B057 AA19 BA11 CA08 CA12 CA16 CC02 DA06 DB02 DB09 DC16 DC22 5C022 AA01 AB12 AB17 AC69 5C054 EA01 FC00 FC03 FC14 FC15 GA01 GB01 GB15 HA19 5C087 AA09 AA23 AA32 AA44 DD05 DD23 EE08 EE16 FE16 FF16 GG16 GG02 GG32 GG08 GG03 GA51 JA11 JA18

Claims (15)

[Claims] 1. A mask detection device provided in a surveillance camera device for photographing a surveillance region, wherein a number of edges extracted from an image photographed by an image pickup portion and an image signal level output from the image pickup portion are predetermined. A gain value for adjusting to a level is input, and when the difference between the number of edges and the reference number of edges is less than a predetermined threshold value and the gain value is equal to or greater than a reference gain value, the monitoring is performed. A mask detection device, characterized in that it is determined that a mask has been applied to a camera device. 2. The mask detection apparatus according to claim 1, wherein the mask determination is performed when the number of edges is less than a predetermined threshold value. 3. The illuminance of the monitoring area is further input,
The mask detection device according to claim 1, wherein the mask determination is performed when the illuminance is equal to or higher than a predetermined threshold value. 4. A mask detection device provided in a surveillance camera device for photographing a surveillance area, wherein a gain value for adjusting an image signal level output from an imaging section to a predetermined level is input, and the first reference When the difference between the gain value and the gain value is a predetermined threshold value or more and the absolute value of the difference between the gain value and the second reference gain value is a predetermined threshold value or more, the surveillance camera device A mask detection device characterized in that it is determined that a mask has been applied to the mask. 5. The illuminance of the monitoring area and the number of edges extracted from the image captured by the imaging unit are input, the illuminance is equal to or more than a predetermined threshold, and the number of edges is predetermined. The mask detection device of the surveillance camera device according to claim 4, wherein the mask determination is performed when the value is less than a threshold value. 6. A mask detection device provided in a surveillance camera device for photographing a surveillance region, wherein the illuminance of the surveillance region and a gain value for adjusting the image signal level output from the imaging unit to a predetermined level. When,
Is input, and when the illuminance is equal to or higher than a predetermined threshold value and the gain value is equal to or higher than a predetermined threshold value, it is determined that the surveillance camera device is masked. 7. The mask determination according to claim 6, further comprising inputting the number of edges extracted from the image captured by the imaging unit, and performing the mask determination when the number of edges is less than a predetermined threshold value. Mask detector. 8. A mask detection device provided in a surveillance camera device for photographing a surveillance region, wherein the illuminance of the surveillance region and an aperture value of an imaging unit are input, and the illuminance is less than a predetermined threshold value. The mask detection device is characterized in that it is determined that the surveillance camera device is masked when the aperture value is equal to or larger than a predetermined threshold value. 9. A mask detection device provided in a surveillance camera device for photographing a surveillance region, wherein the average brightness of the entire image extracted from the image photographed by the imaging unit, the average of each of the blocks obtained by dividing the image into a plurality of blocks. Luminance and a gain value for adjusting the image signal level output from the imaging unit to a predetermined level are input, the gain value is a maximum value, and the average brightness of the entire is less than a predetermined threshold value. The mask detection device is characterized in that when the number of blocks in which the average brightness exceeds a predetermined threshold value is equal to or larger than a predetermined threshold value, it is determined that the surveillance camera device is masked. 10. The mask detection device according to claim 9, further comprising inputting an illuminance of the monitoring area, and performing the mask determination when the illuminance is equal to or more than a predetermined threshold value. 11. A mask detection device provided in a surveillance camera device for photographing a surveillance region, wherein the illuminance of the surveillance region and a shutter speed of an imaging unit are input, and the illuminance is equal to or greater than a predetermined threshold value. The mask detecting device is characterized in that it is determined that the surveillance camera device is masked when the shutter speed value is less than a predetermined threshold value. 12. A mask detection device provided in a surveillance camera device for photographing a surveillance region, wherein the illuminance of the surveillance region, the number of edges extracted from the image photographed by the image pickup unit, and the output from the image pickup unit. The gain value for adjusting the image signal level to a predetermined level and the average brightness of each of the blocks obtained by dividing the image extracted from the image captured by the image capturing unit are input, and the illuminance is set to a predetermined value. It is equal to or more than the threshold value, and the gain is 0.
And the number of edges is less than a predetermined threshold,
The mask detection apparatus, wherein when the difference between the maximum value and the minimum value of the average brightness of the plurality of blocks is less than a predetermined threshold value, it is determined that the surveillance camera device is masked. 13. A surveillance camera device comprising at least one of the mask detection devices according to claim 1. Description: 14. The surveillance camera apparatus according to claim 13, further comprising an output unit that issues a mask abnormality when one of the mask detection apparatuses determines that the mask is continuously masked for a predetermined time. 15. The value of each reference is updated when an image captured by the image capturing unit changes normally.
The mask detection device according to any one of claims 1 to 5.