JP2007318480A - Monitoring interference suppressing device and monitoring interference suppressing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a monitoring interference suppressing device for preventing a video impersonation attack by distinguishing a camera failure from the video impersonation. <P>SOLUTION: The monitoring interference suppressing device 2,000 is provided with a camera having a photographing part 201 for imaging a monitoring target 11, a video scene change detecting part 207 for detecting video scene change of a video photographed by the photographing part 201, a verification video generating part 202 for generating temporal change of video information for verifying whether to photograph the monitoring target 11, a camera failure detecting part 206 for detecting a failure of the camera 2, and an impersonation detecting part 204 for detecting an impersonation attack to the camera 2 when comparing the temporal changes between video information generated by the verification video generating part 202 and video information of a video obtained by photographing the monitoring target 11 by the photographing part 201 to find a large difference, and can distinguish a failure of the camera 2 from video impersonation to detect the impersonation. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to security (protection against impersonation attacks) in a monitoring system.

  As a countermeasure for suppressing interference with conventional monitoring, Patent Document 1 discloses a technique for physically protecting a monitoring system with a camera cover or the like. Patent Document 2 discloses a technique for detecting an abnormality in a transmission line and a power supply and backing up an attack on the transmission line and the power supply of the monitoring system. Japanese Patent Application Laid-Open No. H10-228707 discloses a technique for performing impact detection for detecting a physical attack on a camera. Japanese Patent Application Laid-Open No. H10-228688 discloses a technique for detecting shooting interference by determining the brightness of a shooting environment.

  From these techniques, general defense methods against physical attacks on the camera are disclosed.

  As a conventional technique for video “spoofing” in a monitoring system, the monitoring device receives an operation instruction within a range in which the state information can be collected, and both the state information to be monitored and the state information to be changed by the state change unit are included. A technique for detecting impersonation of video by collecting and analyzing is disclosed in Patent Document 5.

Further, Patent Document 6 discloses a method of implementing countermeasures by using auxiliary lighting or a turning device built in a camera after an obstructing work is detected.
JP 11-372562 A JP-A-62-113781 Japanese Patent Laid-Open No. 2-292777 Japanese Patent Laid-Open No. 11-139471 JP 2005-333415 A JP 2001-6056 A

  In the conventional technology, a method that allows continuous shooting with physical protection and backup against camera shooting obstruction, an impact detection function for physical camera attacks, and image processing of shooting targets Thus, a method for detecting shooting interference has been proposed. In such a conventional technique, a phenomenon in which the camera does not operate normally (hereinafter referred to as a camera failure) due to a malfunction of the monitoring system software or hardware is detected as an image impersonation, not an image impersonation. In addition, when the monitoring system is used for a long period of time, it is known that the monitoring system is frequently reset due to, for example, a software defect.

  However, in the prior art, it is not possible to distinguish between camera failure and video impersonation. Since the two cannot be distinguished from each other, the countermeasures for each problem are delayed, and the burden on the surveillance camera administrator is heavy.

  The present invention solves the problem that it is impossible to distinguish between the above-mentioned prior art image spoofing and camera failure, and for the purpose of improving security in the surveillance system, it is possible to distinguish between camera failure and image spoofing. An object of the present invention is to provide a monitoring interference suppression device having a suppression function.

  The present invention provides a camera having a photographing unit for photographing a video to be monitored, a verification video generating unit for generating a temporal change in video information for verifying whether the monitoring target is being photographed, and detecting a camera failure If there is a difference between the video information generated by the camera failure detection unit and the verification video generation unit and the temporal change of the video information of the video captured by the imaging unit, the impersonation attack on the camera By the monitoring interference suppression device provided with the spoofing detection unit for detecting that it is, it is possible to distinguish and detect spoofing and camera failure in the monitoring device.

  According to the surveillance interference suppression device of the present invention, it is possible to provide a surveillance interference suppression device that can distinguish between surveillance camera failure and video impersonation by having a camera failure detection unit having a function of detecting camera failure.

  Embodiments of the present invention will be described below with reference to the drawings.

  First, referring to FIG. 1, a description will be given of photographing interference caused by “spoofing” in the conventional surveillance interference suppression system 1000.

  In FIG. 1, two examples of shooting disturbance caused by image spoofing will be described. The first example of spoofing is a transmission line 102 that transmits a video of the monitoring target 11 from the camera 100 that is observing the monitoring target 11 to the video receiving device 101. The monitoring target 11 is stolen by exchanging the image with a normal image) and obstructing photographing (case 1 shown in FIG. 1). Specifically, a relay device (not shown) that relays the transmission path 102 forcibly switches to a different video source, or temporarily destroys the transmission path 102 to forcibly switch to a different video source. It can be realized by switching. The second example of spoofing is to cover the camera 100 with a partition 103 that has captured a normal image, and to cause the camera 100 to capture the partition 103 that has displayed a normal image, to disturb the shooting, and to monitor 11 is stolen (case 2 shown in FIG. 1). Specifically, the screen 103 sent from the camera 100 to the video receiving apparatus 101 is replaced with a normal video by pasting and covering the screen 103 displaying the normal video on the camera 100.

(Embodiment 1)
FIG. 2 shows the surveillance interference suppression apparatus 2000 of the present invention that detects the video spoofing by distinguishing the video spoofing described above from a camera failure.

  The surveillance interference suppression apparatus 2000 includes an imaging unit 201 that captures an image, an image scene change detection unit 207 that detects an image scene change of an image captured by the imaging unit 201, and verification that generates image information for verifying impersonation. The video generation unit 202, the camera control unit 203 that controls the imaging unit 201 and the verification video generation unit 202, the temporal change of video information captured by the imaging unit 201, and the temporal change of video information generated by the verification video generation unit 202 If there is a difference between them, an impersonation detection unit 204 that detects impersonation, an output unit 205 that displays an output from the impersonation detection unit 204, and a camera that detects a camera failure from the aspect of software or hardware The failure detection unit 206 is configured.

  Here, the video indicates a continuous image captured by the imaging unit 201, and the video scene is defined as an image at a certain time in the video captured by the imaging unit 201. When impersonation is detected, the impersonation detection unit 204 causes the output unit 205 to display the detection result superimposed on the captured image.

  Note that the impersonation detection unit 204 may be provided inside the relay unit or the output unit 205 in the middle of transmitting video from the camera 2. Specifically, the transmission method may be analog or digital. The transmission medium may also be wired or wireless. The transmission may or may not be relayed. The output unit 205 is a display for a TV or a computer.

  Here, FIG. 8 shows a flowchart of the monitoring disturbance device 2000. FIG. 8 shows an outline of the operation of the monitoring interference suppression apparatus 2000, which is roughly divided into three steps. This will be described below.

  In step 001, it is determined whether the video scene has changed over time. In this step, it is determined whether or not a change has occurred in the image captured by the image capturing unit 201. For example, the video scene changes before and after a change (passing a person or the monitoring target 11 moves) occurs when a person passes or the monitoring target 11 moves due to some kind of trigger. It is determined whether or not this change has occurred (a specific method will be described later). If it is determined that a change in the video scene has occurred, the operation of step 002 is performed. If it is determined that the video scene has not changed, the process returns to the start.

  In step 002, the camera failure detection unit 206 determines whether the camera 2 is out of order (a specific method will be described later). When it is determined that there is a camera failure, the camera failure detection unit 206 displays the camera failure on the output unit 205. If it is determined that the camera is not malfunctioning, the operation of step 003 is performed.

  In step 003, it is determined whether or not the video is impersonated (a specific method will be described later).

  If it is determined in step 003 that the image is impersonated, the impersonation detection unit 204 displays the image impersonation on the output unit 205. If it is determined that the image is not impersonated, the process returns to the start. Note that steps 002 and 003 may be reversed.

  The details of the operation from step 001 to 003 will be described below.

(Determination of whether the video scene has changed (Step 001 in FIG. 8))
The video scene change detection unit 207 determines the change of the video scene (image at a certain time in the video). The determination method of the video scene change is determined from the time change for each video scene of luminance and color information from the video. Specifically, first, a video scene is divided, and a luminance and color histogram for each unit of the divided video scene is calculated. Thereby, each feature amount of the divided video scene can be collected, and the change of the video scene can be detected by observing the time change of each divided video scene. When the time change of the divided video scene has occurred, it is determined that the video scene has changed. Such a method can be realized by using a motion vector detection function standardized in a video encoding method such as MPEG.

(Determining whether the camera is broken (step 002 in FIG. 8))
A method for determining whether or not the camera failure detection unit 206 in step 002 of FIG. First, what kind of phenomenon occurs in the failure of the camera 2 will be described.

  As a phenomenon that appears due to a failure of the camera 2, symptoms such as “reset” or “freeze” of the camera 2 due to software or hardware malfunctions can be given. When the camera 2 is reset, the camera 2 is restarted or the captured video is disturbed. Further, when the camera 2 freezes, the camera 2 does not accept a new control instruction or the captured video does not change.

  Hereinafter, a phenomenon that occurs in the case of “reset” and a method for detecting the phenomenon will be described.

  In the case of a reset caused by software, for example, there is unauthorized access to the memory. In the unauthorized access to the memory, the operation of the camera 2 is initialized by unauthorized access to an unexpected memory location. In general, unauthorized access can be detected by notifying an error or the like when moving to a program that handles exception processing by the OS. When this error is notified, it can be detected that the reset is caused by software.

  In the case of reset caused by hardware, for example, a cable contact failure is given. The cable contact failure can be detected by observing whether the cable signal line is periodically disconnected. Specifically, when the voltage of the signal line of the cable is observed and the voltage changes with time (for example, the voltage when the connector is coming off and operating normally and the other voltage are If it repeats in time), it can be detected that the reset is caused by hardware.

  Hereinafter, a phenomenon that occurs in the case of “freeze” and a method for detecting the phenomenon will be described.

  In the case of freeze caused by software, for example, there is a deadlock due to a failure in communication between software processing processes. Here, the deadlock is a state in which processing processes that are simultaneously executing each lock necessary resources and do not release the locks unless they complete each other. The elapsed time of the processing time of the processing process is measured, and when a certain time or more has elapsed, it can be detected that the freeze is caused by software (a deadlock has occurred).

  In the case of freeze due to hardware, for example, there is a break in the cable. The disconnection of the cable can be detected by observing whether the disconnection of the cable signal line continues for a certain time or more. Specifically, the cable line that transmits the operation instruction of the camera 2 may be broken. By observing the voltage of the signal line in the same manner as the hardware reset, it is possible to detect that the hardware is causing the freeze.

  FIG. 4 is a flowchart of an operation in which the camera failure detection unit 206 determines a failure of the camera 2. The camera failure detection unit 206 checks the cause of the camera 2 failure from the viewpoint of reset or freeze as described above. As for reset, it is determined by the above-described method whether the cause of the reset is hardware or software. Similarly, regarding the freeze, it is determined whether the cause of the freeze is in hardware or software by the above-described method.

  If the camera failure detection unit 206 determines that the camera is neither reset nor freeze, it is determined that the camera 2 is not malfunctioning, and a video impersonation determination process (step 003 in FIG. 8, a specific method will be described later) is performed. As described above, according to the present invention, since the camera failure detection unit 206 is included, it is possible to determine the failure of the camera 2, so that it is possible to detect the image spoofing by distinguishing the image spoofing from the camera 2 failure. As a result, the surveillance staff can respond to each of them quickly without impersonating whether the camera 2 is faulty or not.

  Note that, as described above, even if the operation of the camera failure detection unit 206 (step 002 shown in FIG. 8) is performed after the operation of the impersonation detection unit 204 (step 003 shown in FIG. 8), it is an effect of the present invention. Image spoofing can be detected by distinguishing image spoofing from camera failure.

(Determination of video impersonation (step 003 in FIG. 8))
6 is a flowchart for determining impersonation in FIG. 5. If the camera failure unit 206 determines in step 002 in FIG. 8 that the camera 2 has not failed, the operation from the start in FIG. 5 is performed.

  The verification video generation unit 202 transmits an instruction (camera parameters such as pan, tilt, and zoom) to operate the camera 2 at random to the camera control unit 203, and the camera control unit 203 operates the imaging unit 201 based on the operation instruction. (Step 301). The parameters for operating the photographing unit 201 of the camera 2 may use information such as the exposure time and white balance of the camera 2 instead of the camera parameters, or may be used in combination with each other.

  Next, the camera 2 transmits the video image of the monitoring target 11 to the spoofing detection unit 204 (step 302).

  The verification video generation unit 202 uses video information (camera parameters such as pan, tilt, and zoom) that randomly operates the photographing unit 201 of the camera 2 to obtain video information (luminance for each divided video unit estimated from the camera parameters, or A time change of color information is generated (step 303).

  The impersonation detection unit 204 generates a time change of the video information generated from the video received from the camera 2 (the video scene is divided and the luminance or color information for each unit of the divided video scene) (step 304). .

  The video information created by the impersonation detection unit 204 is compared with the temporal change of the video information generated by the verification video generation unit 202 (step 305). Here, the impersonation detection unit 204 compares the video information generated from the video received from the camera 2 with the video information generated by the verification video generation unit 202, and if the same, the imaging unit 201 captures a normal video. If there is a difference, it is determined that the shooting of the shooting unit 201 is impersonating the video. Note that the time change determination method of video information can be realized by using a motion vector detection function standardized by a video encoding method such as MPEG as in the above-described determination of a video scene change. Specifically, if a large difference in motion vectors continues to occur, it is determined that video impersonation has occurred.

  If there is no difference between the time changes of both pieces of video information, the verification video generation unit 202 is stopped (step 306).

(Modification of Embodiment 1)
FIG. 6 is a diagram for explaining a monitoring disturbance suppression system 2001 as a modification of impersonation detection. The surveillance interference suppression system 2001 includes an imaging unit 201 that captures an image, an image scene change detection unit 207 that detects an image scene change of an image captured by the imaging unit 201, and a verification image that generates an image for verifying impersonation. From the generation unit 202, the shooting target specifying unit 208 that specifies the shooting target 209, the camera control unit 201 that controls the shooting target specifying unit 208, and the failure detection unit 301 that detects a failure of the camera 2 When there is a difference between the configured camera 2 and the video information generated from the video output from the camera 2 and the video information generated by the verification video generation unit 202, an impersonation detection unit 204 that detects impersonation The output unit 205 displays the output from the impersonation detection unit 204. When impersonation is detected, the impersonation detection unit 204 causes the display unit 205 to display the detection result (image impersonation) superimposed on the captured image. Note that the spoofing detection unit 204 may be provided in the repeater or the output unit 205 in the middle of video transmission. In the surveillance interference suppression system 2001, the camera failure detection unit 206 of the surveillance interference suppression device 2000 of Embodiment 1 is replaced with the failure detection unit 301, and a shooting target 208 is added.

  In the operation of FIG. 8, the monitoring interference suppression system 2001 performs the same operation as that of the monitoring interference suppression device 2000 in steps 001 and 002, and step 003 performs an operation different from that of the monitoring interference suppression device 2000. FIG. 7 shows a flowchart of the operation of step 003 in FIG.

  The shooting target specifying unit 208 specifies the shooting target 209 for the shooting target 11, and transmits information of the shooting target 209 from the shooting target specifying unit 208 to the verification video generation unit 202 (step 601). Here, the information of the shooting target 209 is transmitted to the verification video generation unit 202 because the different operation (light emission operation (described later) or rotation of the turntable (described later)) corresponding to the shooting target 209 is generated. This is because the unit 202 generates.

  The shooting target specifying unit 208 transmits an operation instruction corresponding to the shooting target 209 of the verification video generation unit 202 to the shooting target special item 209 (step 602).

  Note that the activation timing of the shooting target specifying unit 208 is performed at the timing when the change of the video scene occurs, as described above.

  Next, the shooting target 209 receives transmission information from the shooting target specifying unit 208, and based on the received information, emits a specific pattern of visible light from the shooting target 209, or places the monitoring target 11 on the turntable. Then, the turntable is operated (step 603). Specifically, the principle that the TV channel changes when the TV is controlled by the remote controller is applied, and according to the selection of the remote controller from the monitoring target 11 according to the infrared light emitted from the remote controller (corresponding to the shooting target specifying unit 208). By emitting visible light at predetermined intervals, or by rotating the base in advance according to the selection of the remote controller, it is possible to easily generate various patterns of images. The following operation of the impersonation detection unit 204 is the same as that of the first embodiment.

  Note that the failure detection unit 301 may also detect a failure in a portion of the verification video generation unit 202 that realizes an operation instruction different from the first embodiment in addition to the failure of the camera 2. Specifically, a light receiving unit that receives information from the shooting target specifying unit 208 of the shooting target 209, a light emitting unit that emits a specific pattern of visible light, and a motor that operates the turntable are detected. .

  As described above, by including the failure detection unit 301, it is possible to detect impersonation by distinguishing impersonation as a failure.

  The monitoring interference suppression device and the monitoring interference suppression system according to the present invention are useful for monitoring that requires high security because they can distinguish between failure and video impersonation, and are used for monitoring jewelry, art objects, safes, etc. Not only can it be used for monitoring people at entrances and exits of facilities and elevators.

Diagram explaining shooting disturbance caused by "spoofing" in a conventional surveillance system The figure explaining the basic composition of surveillance disturbance control device 2000 Diagram explaining failure due to software and hardware The figure explaining operation | movement of the camera failure detection part 206 Flowchart explaining operation of impersonation detection The figure explaining the monitoring interference suppression system 2001 which can identify the imaging | photography object 11. FIG. The flowchart explaining operation | movement of the monitoring interference suppression system 2001 which can specify the imaging | photography object 11. FIG. The flowchart explaining the outline | summary of operation | movement of the monitoring interference suppression apparatus 2000 and the monitoring interference suppression system 2001.

Explanation of symbols

2 Camera 11 Monitored Object 101 Video Receiver 102 Transmission Path 103 Partition 201 Shooting Unit 202 Verification Video Generation Unit 203 Camera Control Unit 204 Impersonation Detection Unit 205 Output Unit 206 Camera Failure Detection Unit 207 Video Scene Change Detection Unit 208 Shooting Target Object Specification Unit 209 Shooting target 301 Failure detection unit 1000 Monitoring interference suppression system 2000 Monitoring interference suppression device 2001 Monitoring interference suppression system

Claims (6)

A camera having an imaging unit that captures a video to be monitored, a verification video generation unit that generates a temporal change in video information for verifying whether the target is being captured, and a camera failure detection that detects a camera failure If there is a difference between the time change of the video information generated by the video and the verification video generation unit and the time change of the video information of the video captured by the imaging unit, the spoofing attack to the camera A monitoring jamming suppression device provided with a spoofing detection unit for detecting that it is. The monitoring disturbance suppression apparatus according to claim 1, wherein the video scene change detection unit that detects a change in the video scene and the verification video generation unit are operated based on a result of the video scene change detection unit. The monitoring interference suppression apparatus according to claim 1, further comprising the verification video generation unit configured to generate a video for verification by operating a camera. 2. The monitoring disturbance suppression device according to claim 1, wherein the operation parameter of the camera is compared with an image captured by the camera, and if there is no difference, the operation of the camera is stopped. Based on transmission information transmitted from the shooting target specifying unit and a shooting target specifying unit for specifying a shooting target for the shooting target, the shooting target responds with information based on the transmission information, or The monitoring disturbance suppression apparatus according to claim 1, further comprising the verification video generation unit that operates a shooting target. A shooting unit shooting a video to be monitored;
A step in which a verification video generation unit generates a change in video information for verifying whether the camera is shooting a monitoring target;
A camera failure detection unit detecting a camera failure;
A step of detecting an impersonation in which an impersonation detection unit detects an impersonation attack on the camera by comparing a temporal change of the video information generated by the verification video with a temporal change of the video information of the video captured by the imaging unit; And a monitoring disturbance suppression method.

Images