JP2011004050A - Monitoring camera system and recorder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a monitoring camera system and a recorder which identifies a suspicious person from a video signal output from a camera, in a short time.SOLUTION: When face detection data including a direction of a face are added to a received video signal, the number of changes in the direction of the face is counted by the face detection data, and videos and the number of changes are recorded, and a video including the changes equal to or higher than a predetermined value is extracted from a recording part. Consequently, a suspicious person can be identified in a short time.

Description

  The present invention relates to a surveillance camera system and a recording apparatus.

  The surveillance camera system mainly includes a camera that images a subject and outputs a video signal, a turntable that rotates the camera in a pan direction and a tilt direction, a recorder that records a video signal output from the camera, and a camera. It is comprised from the monitor which displays the video signal output or the video signal recorded on the recorder.

  Some surveillance camera systems have a function of identifying a suspicious person. For example, a suspicious person notification system disclosed in Patent Document 1 includes an image detection unit that detects a face image of a visitor from a video signal output from a camera, and an imaging database that stores entrance data of the visitor including the face image. A resident database in which resident data including a face image is registered, face authentication means for determining whether the resident is a resident from the admission data and the resident data, non-resident admittance data, and predetermined data Suspicious person determination means for determining whether or not a non-resident is a suspicious person from suspicious action conditions, a management center for identifying a suspicious person from suspicious person data including a face image, suspicious person data including a face image, and A suspicious person judging means for judging whether or not the non-resident is a suspicious person from the admission data of the non-resident, and the suspicious person judging means is based on the number of times the non-resident has entered the face determined by the face authentication means. , It determines whether or not the non-resident is a suspicious behavior person.

JP 2006-120084 A

  However, when identifying a suspicious person using face information detected from a video signal output from a camera, a method of performing face authentication by comparing the detected face information is common, but the accuracy of face authentication is It is not expensive at all, and it takes time for face recognition, and a suspicious person cannot be identified in a short time. In particular, in the technique disclosed in Patent Document 1 described above, whether or not the non-resident is a suspicious person is determined based on the number of times the non-resident enters the building, which is determined by the face authentication means. It is not possible to identify a suspicious person in a short time.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a surveillance camera system and a recording apparatus that can identify a suspicious person in a short time from a video signal output from a camera.

  The surveillance camera system of the present invention detects a human face from a captured video, outputs a face detection data including at least the face direction together with a video signal, and receives a video signal output from the camera. A control unit; a face detection unit for determining whether the face detection data is added to the video signal received by the NW control unit; and the face detection data is added to the video signal by the face detection unit. A recording unit having a counter unit that counts the number of times the face orientation has changed, and a recording unit that records the video from the camera and the number of times. .

  According to the above configuration, when face detection data including the face direction is added to the received video signal, the number of times the face direction is changed from the face detection data is counted, and the video and the number of times are recorded. Suspicious person can be identified by time.

  In the above configuration, the recording apparatus extracts the video from the camera whose number of times is equal to or greater than a predetermined value from the recording unit.

  According to the above configuration, since the video whose number of face orientation changes is equal to or greater than a predetermined value is extracted from the recording unit, it is possible to identify a suspicious person in a short time.

  The recording apparatus of the present invention includes an NW control unit that receives a video signal output from a camera, a face detection unit that determines whether face detection data is attached to the video signal received by the NW control unit, When the face detection unit determines that the face detection data is added to the video signal, the counter unit counts the number of times the face direction has changed, and records the video from the camera and the number of times. And a recording unit, wherein the video from the camera having the number of times equal to or greater than a predetermined value is extracted from the recording unit.

  According to the above configuration, when face detection data including the face direction is added to the received video signal, the number of times the face direction is changed from the face detection data is counted, and the video and the number of times are recorded. Since a video having a value equal to or greater than a predetermined value is extracted from the recording unit, a suspicious person can be identified in a short time.

  The recording method of the present invention includes a step of receiving a video signal output from a camera, a step of determining whether face detection data is attached to the received video signal, and the face detection data being attached to the video signal. If it is determined that the orientation of the face has been changed, the method includes a step of counting the number of times the face orientation has changed, and a step of recording the video from the camera and the number of times, wherein the number of times is a predetermined value or more. Extracting video from the camera from the recording unit.

  According to the above method, when face detection data including the face direction is added to the received video signal, the number of times the face direction is changed from the face detection data is counted, the video and the number of times are recorded, and the number of times Since a video having a value equal to or greater than a predetermined value is extracted from the recording unit, a suspicious person can be identified in a short time.

  In the present invention, when face detection data including the face direction is added to the received video signal, the number of times the face direction is changed from the face detection data is counted, and the video and the number of times are recorded. Suspicious person can be identified.

The block diagram which shows schematic structure of the surveillance camera system which concerns on one embodiment of this invention Schematic diagram showing face detection data transmitted from the camera to the recorder in the surveillance camera system according to the present embodiment The flowchart which shows the operation | movement at the time of the video signal recording containing the user operation of the surveillance camera system which concerns on this Embodiment The flowchart which shows the operation | movement at the time of a suspicious person search including the user operation of the surveillance camera system which concerns on this Embodiment

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments for carrying out the invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing a schematic configuration of a surveillance camera system according to an embodiment of the present invention. In the figure, a surveillance camera system 1 according to the present embodiment includes a plurality (three in the figure) of cameras 2, a recorder (recording device) 3 that records video signals output from each of the plurality of cameras 2, and a plurality of cameras 2. HUB (concentrator) 4 for connecting the camera 2 to the recorder 3, a browser 5 for controlling the HUB 4 to perform camera selection, an area for detecting a suspicious person for the recorder 3, And a browser 6 for setting a period for searching for a suspicious person.

In addition to the video signal, the camera 2 transmits face detection data including the following information to the recorder 3.
(1) Face horizontal / vertical start position (2) Face horizontal / vertical width (3) Face orientation (front, right / left 30 degrees, 45 degrees, 75 degrees)

  FIG. 2 is a schematic diagram showing face detection data (face information) transmitted from the camera 2 to the recorder 3. In the figure, “SP” is the face horizontal / vertical start position, “HW” is the face horizontal width, “VW” is the face vertical width, and “FI” is the face image.

The seven types of information on the face orientation in (3) are represented by 7 bits, for example.
Front: 1000000
Right 30 degrees: 0100000
30 degrees left: 0010000
Right 45 degrees: 0001000
45 degrees left: 0000100
Right 75 degrees: 0000010
75 degrees left: 0000001

  The information of (1) to (3) is transmitted for a maximum of eight people (information of (1) to (3) is given one by one).

  The recorder 3 has a function of identifying a suspicious person using the face detection data transmitted from the camera 2. As shown in FIG. 1, the recorder 3 includes an NW (Net-Work) control unit 31, a memory 32, a writing / reading unit 33, a recording unit 34, a file system unit 35, and a timer unit 36. I have. The NW control unit 31 controls panning, tilting, zooming, and the like of each camera 2, takes in video signals (including video signals with face detection data added) from each camera 2, controls the file system unit 35, and timers The part 36 is controlled. Further, the NW control unit 31 sets an area for detecting a suspicious person (hereinafter referred to as “suspicious person detection area”) in accordance with an instruction from the browser 6. A region indicated by reference numeral 8 in FIG. 2 is a suspicious person detection region. The memory 32 is a volatile memory such as a RAM, and temporarily records a video signal. The writing / reading unit 33 writes the video signal recorded in the memory 32 to the recording unit 34 and reads the video signal recorded in the recording unit 34. The recording unit 34 includes a storage device such as a hard disk drive (HDD) and records video signals.

  The file system unit 35 includes a face detection unit 351, an area determination unit 352, a memory 353, and a counter unit 354. The face detection unit 351 determines whether face detection data is attached to the video signal recorded in the memory 32. In the case of a video signal to which face detection data is added, the area determination unit 352 determines whether face detection data is included in the suspicious person detection area 8 set by the NW control unit 31 in accordance with an instruction from the browser 6. When face detection data is included in the suspicious person detection area, a value indicating the face direction included in the face detection data is recorded in the memory 353. In addition, the area determination unit 352 notifies the counter unit 354 that a value indicating the face orientation has been recorded in the memory 353.

  The memory 353 is a nonvolatile memory such as a RAM, and stores a value indicating the face orientation included in the face detection data as described above. When face detection data is included in the suspicious person detection area 8, the counter unit 354 counts the number of times the face orientation has changed within 30 seconds after the initialization process. That is, the number of times the face orientation has changed within 30 seconds after the initialization process is counted. The NW control unit 31 identifies a person having the face as a suspicious person when the number of times the face orientation has changed is equal to or greater than a certain value within 30 seconds. Note that the timer value of the timer unit 36 can be changed. The counter unit 354 and the memory 353 are initialized by the NW control unit 31 when the video signal is recorded by the recorder 3.

  The timer unit 36 counts for 30 seconds, starts a count operation in accordance with a count instruction from the NW control unit 31, and outputs a count value to the counter unit 354 and the NW control unit 31.

  Next, the operation of the surveillance camera system 1 of the present embodiment will be described. FIG. 3 is a flowchart showing an operation at the time of recording a video signal of the surveillance camera system 1 including a user operation. In the figure, first, the user performs an operation for setting a suspicious person detection area, that is, a suspicious person detection area, using the browser 6, and the NW control unit 31 performs the suspicious person detection area 8 (see FIG. 2) according to this operation. ) Is set (step S10). After setting the suspicious person detection area 8, the NW control unit 31 performs an initialization process (step S11). That is, the counter unit 354 that counts the number of times the face orientation has changed is initialized, and the memory 353 that stores the value indicating the previous face orientation (face orientation value) is initialized. In addition, the NW control unit 31 gives a count start instruction to the timer unit 36 after the initialization processing of the counter unit 354 and the memory 353, and starts the 30-second timer of the timer unit 36.

  After starting the 30-second timer, the NW control unit 31 determines whether a video signal from the camera 2 has been received (step S12). If the video signal from the camera 2 is not received, this step S12 is repeated until it is received. When the video signal from the camera 2 is received, it is determined whether 30 seconds have elapsed (step S13). When 30 seconds have elapsed, the counter unit 354 and the memory 353 are initialized again, and at the same time, the 30-second timer is started (step S14). If 30 seconds have not elapsed, the process of step S14 is omitted and the process proceeds to the next step.

  When 30 seconds have not elapsed, or after 30 seconds have elapsed, the counter unit 354 and the memory 353 are initialized and the 30-second timer is restarted, and then the video signal received from the camera 2 by the face detection unit 351 It is determined whether or not there is face detection data (step S15). If there is face detection data, the area determination unit 352 determines whether the face detection data position is included in the suspicious person detection area 8 (step S16). When the face detection data position is included in the suspicious person detection area 8, the area determination unit 352 determines whether the face orientation of the person is different from the previous one (step S17). If the face orientation of the person is different from the previous one, the value of the counter unit 354 is incremented by 1, and the face orientation value is recorded in the memory 353 (step S18). After these processes are performed, the writing / reading unit 33 records the face detection data, the recording time, the value of the face direction counter, and the face direction value received from the camera 2 in the recording unit 34.

  If there is no face detection data in the video signal received in step S15, or if the face detection data position is not included in the suspicious person detection area 8 in step S16, or if the face orientation of the person has been previously determined in step S17 If it is not different (ie, the same), the process proceeds to step S19. After finishing the process of step S19, it returns to step S12 and repeats the process of each above-mentioned step.

  FIG. 4 is a flowchart showing an operation at the time of searching for a suspicious person of the surveillance camera system 1 including a user operation. In the figure, first, the user performs an operation for setting a period for searching for a suspicious person using the browser 6, and further performs an operation for instructing execution of the search (step S30). When a period for searching for a suspicious person is set and search execution is instructed, the NW control unit 31 operates in cooperation with the writing / reading unit 33, and the value of the face direction counter within the set period. A part where the value (the value obtained by counting the number of changes in the face direction and recorded in the recording unit 34) is equal to or greater than a certain value (the certain value can be changed) is searched (step S31). That is, the video whose face direction counter value is equal to or greater than a certain value is extracted from the recording unit 34. If there is a corresponding part, it is specified that the person in the image at that part is a suspicious person (step S32).

  As described above, according to the surveillance camera system 1 of the present embodiment, when face detection data including the face direction is added to the video signal output from the camera 2, the number of times the face direction is changed is counted. However, when the number of times is equal to or greater than a certain value within 30 seconds, the person having the face is identified as a suspicious person, and therefore, the suspicious person can be identified in a short time.

  In this embodiment, the suspicious person is specified by determining the change in the orientation of the face. However, the suspicious person may be specified by determining the change in the movement of the eyes, A suspicious person may be specified by determining both a change in face orientation and a change in eye movement.

  The present invention has an effect that a suspicious person can be specified in a short time, and can be applied to a surveillance camera system.

1 surveillance camera system 2 camera 3 recorder 4 HUB
5, 6 Browser 31 NW control unit 32, 353 Memory 33 Write / read unit 34 Recording unit 35 File system unit 36 Timer unit 351 Face detection unit 352 Area determination unit 354 Counter unit

Claims (4)

A camera that detects a person's face from the captured video and outputs face detection data including at least the face direction together with a video signal;
An NW control unit that receives a video signal output from the camera, a face detection unit that determines whether the face detection data is added to the video signal received by the NW control unit, and the face detection unit When it is determined that the face detection data is attached to the video signal, the counter includes a counter unit that counts the number of times the face direction has changed, and a recording unit that records the video from the camera and the number of times. A recording device;
A surveillance camera system characterized by comprising:   The surveillance camera system according to claim 1, wherein the recording device extracts an image from the camera in which the number of times is equal to or greater than a predetermined value from the recording unit.   An NW control unit that receives a video signal output from the camera, a face detection unit that determines whether or not face detection data is added to the video signal received by the NW control unit, and the video signal at the face detection unit A counter unit that counts the number of times the direction of the face has changed, and a recording unit that records the video from the camera and the number of times, A recording apparatus for extracting video from the camera, the number of times of which is equal to or greater than a predetermined value, from the recording unit.   A step of receiving a video signal output from the camera; a step of determining whether face detection data is attached to the received video signal; and determining that the face detection data is attached to the video signal. The number of times the face orientation has changed, and a step of recording the video from the camera and the number of times, and recording the video from the camera, the number of times being equal to or greater than a predetermined value. And a step of extracting from the recording section.

Images