JP2005286468A - Monitoring system having masking function, camera, mask releasing device used with camera - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a monitoring system for making a mask to be released according to the authority of an observer different even if there are a plurality of observers having a mask release right. <P>SOLUTION: The monitoring system 2 comprises an imaging section 4; a masking section 6 that creates a mask image in which an object is masked when at least one object to be masked is included in an original image captured by the imaging section, and generates mask information for indicating position/size on the original image at a mask region regarding each object and the type of the object; an unmask processor 48 for releasing the mask of a mask image, based on the mask information; a storage 52 that stores a table for indicating the correspondence between the observer who monitors the mask image and the type of the object that can be observed by the observer; and a control unit 50 that refers to the table when a command is received from an observer via an input 54, and controls the unmask processor so that only a mask regarding the type of object corresponding to the observer is released. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention generally relates to a surveillance system and a camera, and more particularly to a surveillance system and a camera having a function of masking a part of an image related to privacy. The present invention also relates to a mask release device used with the surveillance camera.

For example, Patent Document 1 discloses a monitoring camera device that masks and displays a part of an image for privacy protection. In this apparatus, only a specific person who has a password can cancel and set the mask (see paragraph [0061]).
JP 2001-69494 A

  There may be a case where there are not only one person who monitors an image with a part masked but also a plurality of persons. The authority of these multiple observers is usually not the same, so even for an object (subject) of a type that needs to be masked for a certain observer, a more authorized observer observes the object. You may wish to do that. However, in the above-described surveillance camera device, even if there are a plurality of privacy objects in the image, and there are a plurality of masks in the image, the plurality of masks can only be released at once.

  Therefore, an object of the present invention is to change the masks to be released according to the authority of the supervisor even when there are a plurality of supervisors having the mask release right.

In order to achieve the above object, a monitoring system according to the present invention includes:
An image input unit;
When one or more objects to be masked are included in the original image input by the image input unit, a mask image is created by masking the objects, and the position of the mask area for each object on the original image And a masking unit for generating mask information indicating the size and the type of the object,
An unmask processing unit for releasing the mask of the mask image based on the mask information;
A storage unit for storing a table indicating correspondence between a monitor who monitors a mask image and the types of objects which can be observed by the monitor;
When receiving a command from a certain monitor, a control unit that controls the unmask processing unit so as to release only the mask related to the type of object corresponding to the monitor with reference to the table stored in the storage unit;
It is characterized by providing.

  According to the present invention, a specific monitor accesses a mask image created by a masking unit based on an original image input by an image input unit (for example, a camera) (for example, a mask image displayed on a monitor is displayed). In the case of observation), the unmask processing unit unmasks only the types of objects (for example, persons) that can be observed by the monitor. Thus, when there are a plurality of supervisors, the masks to be released can be changed according to the authority of the supervisor.

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

Embodiment 1 FIG.
FIG. 1 shows a first embodiment of a monitoring system according to the present invention. In general, the system 2 includes an imaging unit (camera) 4 that captures an image of a certain monitoring area, and an image (original image) captured by the imaging unit 4 includes one or more objects that should protect privacy. For example, the masking unit 6 that masks the object, the transmission unit 8 that transmits the image data and the like masked by the masking unit 6, the reception unit 10 that receives the image data and the like transmitted from the transmission unit 8, A mask canceling unit 12 that cancels part or all of the mask of an image subjected to masking received by the receiving unit 10 and a display unit 14 that displays an image with part or all of the mask removed.

  The imaging unit 4 includes an optical system 16, an imaging element 18, an A / D conversion unit 20, an imaging control unit 22, and the like. The optical system 16 is a lens mechanism composed of a plurality of lenses having optical axes parallel to a predetermined direction, and forms an image of the monitoring area on the imaging surface of the image sensor 18. The imaging element 18 is a CCD, for example, and converts an optical signal received on the imaging surface into an electrical signal and outputs the electrical signal to the A / D converter 20. The imaging element 18 captures an image in synchronization with a vertical drive (VD) signal from the imaging control unit 22. The A / D conversion unit 20 outputs an electrical signal (analog signal) from the image sensor 18 to the masking unit 6 after analog / digital (A / D) conversion.

  The masking unit 6 includes an image memory 23, an object extraction unit 24, a mask area setting unit 26, and a mask processing unit 28. The image memory 23 is for storing time-series image data sent from the imaging unit 4. The object extraction unit 24 is for creating an image in which an object to be masked is extracted from an image frame at each time. The object to be masked here is written on a moving part such as a specific part of a person such as a person or face, a specific part of a car such as a car or license plate, the whole or part of a building, a signboard, etc. A stationary object such as a character. Which object is to be extracted by the object extraction unit 24 is set in advance (for example, a car or a person is an extraction target, but a train is not an extraction target). May be changed.

  The moving body, for example, captures an imaging area in the absence of the moving body in advance to obtain a background image (FIG. 2), and calculates a difference between the background image and each time-of-day image (FIG. 3) for each pixel. Thus, it is extracted as a change area as shown in FIG. After extracting the moving object, the object extracting unit 24 determines whether each extracted moving object is, for example, a car, a person, or another object. Specifically, the object extraction unit 24 shapes and sizes of models (for example, the car model Mc and the human model Mp) on the image corresponding to the objects (for example, the car C and the person P) to be masked in advance. Is stored (FIG. 5). Then, when the degree of coincidence between the extracted moving object and the model is high, it is determined that the moving object is an object to be masked. Even in the same object, the appearance in the image differs depending on the installation position / orientation of the imaging unit 4, and therefore the shape and size of the model also differ depending on the installation position / orientation of the imaging unit 4. After determining which object the moving object is to be masked, the object extraction unit 24 associates the moving objects at each time to be extracted in a plurality of time-series images as shown in FIG. (For example, it is determined that the moving object extracted for the image frame at a certain time and the moving object extracted for the image frame at the next time are the same object). Information data relating to the position, size, etc. of the extracted object on the image frame at each time is output to the mask area setting unit 26.

  As a specific example of the above processing, for example, Hideto Fujiwara et al., “Vehicle tracking method using texture background difference and silhouette model”, the 6th Image Sensing Symposium Lecture Collection, B1, p17- 22 (2000), Tetsuji Hashita et al., “Outdoor intrusion monitoring system using background difference and motion vector analysis”, Proc. Of the 8th Image Sensing Symposium, A2, p7-12 (2002) Can be illustrated.

  With reference to FIG. 3, for example, a stationary object is extracted by performing character recognition if it is a character L, or by extracting a contour line of a building if it is a building B. Instead, an area to be masked in advance by the system administrator may be set.

  The mask area setting unit 26 sets an area to be actually masked based on information data such as the position and size of the object to be masked output from the object extraction unit 24. The area to be masked is, for example, the area itself extracted by the object extraction unit 24, the size of the extraction area that has been changed by applying a predetermined enlargement ratio or reduction ratio, and a part of the extraction area (the extraction area is a person). In this case, only the face is an area to be masked). For example, if the face is an object to be masked, a circular area including the upper part of the area extracted by the object extraction unit 24 and larger than the upper part is set as the area to be masked. Information data regarding the position, size, etc., of each area to be masked is output to the mask processing unit 28. Information indicating the position and size of each area to be masked on the image and the type of object masked in each area (hereinafter referred to as mask information) is stored in the storage unit 30. Yes. The information stored in the mask information storage unit 30 is, as a specific example, with reference to FIG. 7, with respect to the region 1, the region 2 includes information such as the type and character, the position of the quadrilateral region, the coordinates of the vertex With respect to the area 3, the information such as the position of the polygon is the length of the polygonal area, the coordinates of the vertex, etc. The area 3 is the information such as the length of the polygonal area and the coordinates of the vertex, etc. , The type is a person (face), and information such as the position is the center of the circular region and its diameter. Regarding the areas 3 and 4, information stored in the mask information storage unit 30 is updated over time.

  The mask processing unit 28 rewrites the data of each pixel in the area to be masked set by the mask area setting unit 26 in each image frame stored in the image memory 23 according to a predetermined rule. An image partially masked (mask image) is created. The mask is a reversible reversible mask. The mask image is stored in the mask image storage unit 32.

  The reversible mask process is performed as follows, for example. (1) When the image data is RGB data represented in 256 levels of each color 0 to 255, for example, when each color data of each pixel is represented by 8 bits consisting of “0” and “1”, for example, “ Invert “0” and “1”. In this case, for example, when certain color data of a certain pixel is represented by 00111110, the data is converted to 11010001. Further, each value of the coordinates (X, Y) of each pixel is divided by 8, the remainder of X and the remainder of Y are obtained, and the bits corresponding to the respective values of each color data are exchanged. For example, when the coordinates of the pixel are (100, 125), the remainder is 4 and 5, and the fourth bit and the fifth bit of each color data are switched. Alternatively, the upper 4 bits and the lower 4 bits may be interchanged. (2) Instead of correcting the pixel value, it is also conceivable to replace a pixel having a region to be reversibly masked with another pixel in accordance with a certain rule. In addition, it is possible to combine (1) and (2).

  The monitoring system 2 also includes an encoding circuit 34 that performs encoding such as compression on the mask image data output from the mask image storage unit 32, and the encoded mask image data and mask information storage unit 30 as a transmission unit. A multiplexing circuit 36 that multiplexes the stored mask information data to generate multiplexed data and outputs the multiplexed data to the transmitter 8 is provided.

  The transmission unit 8 and the reception unit 10 are connected via a communication line 38. As the communication line 38, for example, a wired transmission line, a wireless transmission line, or a combination thereof such as a digital telephone line (for example, ISDN), the Internet, a LAN, or a cellular phone line is used. Note that an ADSL system using an analog line may be used as the communication line 38.

  The monitoring system 2 further includes a separation circuit 40 that separates multiplexed data output from the receiving unit 10 into mask image data and mask information data, and a decoding circuit 42 that performs decoding such as decompression on the mask image data. The mask image data and the mask information data are output to the mask image storage unit 44 and the mask information storage unit 46 of the mask release unit 12, respectively.

  In addition to the mask image storage unit 44 and the mask information storage unit 46, the mask release unit 12 includes an unmask processing unit 48, a control unit 50, a supervisor authority information storage unit 52, and an input unit 54.

  The unmask processing unit 48 receives a mask image from the mask image storage unit 44 and releases part or all of the mask of the mask image. The cancellation is performed by returning the pixel converted based on a certain rule in the mask processing unit 28 of the masking unit 6 to the original pixel.

  The supervisor authority information storage unit 52 is for storing the types of objects that can be unmasked according to the authority of the supervisor. The supervisor who monitors the mask image and the types of objects that can be observed by the supervisor For example, a table indicating the correspondence between the supervisor ID number and the object that can be unmasked is stored as shown in FIG. In this example, there are five supervisors, and the assigned ID number S1 is the car only, the S2 supervisor is only the person, the S3 supervisor is the car and the person, and the S4 supervisor is all. The mask for the object is released to enable observation. The supervisor whose ID number is S5 does not have the release right.

  The input unit 54 is, for example, a keyboard or a mouse. When the monitor inputs an ID number and a password displayed on a monitor (not shown) or a monitor 58 (described later) of the display unit 14 via the input unit 54, the control unit 50 Based on the input ID number and the supervisor authority information stored in the supervisor authority information storage unit 52, a signal indicating the type of the object to be unmasked is output to the unmask processing unit 48. When the unmask processing unit 48 receives a signal from the control unit 50, the unmask processing unit 48 stores the mask information stored in the mask information storage unit 46 (the position and size of each mask on the image, and the type of object masked in each region). It comes to refer. The unmask processing unit 48 obtains information on the type of object to be unmasked from the control unit 50, obtains mask information from the mask information storage unit 46, and cancels some or all masks. For example, when the supervisor whose ID number is S3 accesses the system 2 via the input unit 54, the mask that has hidden the car and the person is released.

  When a monitor who does not have the release right accesses the system 2 via the input unit 54, the control unit 50 refers to the input ID number and the monitor authority information storage unit 52, and there is no object to be released. As a result, no mask release command is output to the unmask processing section 48.

  The display unit 14 displays an image memory 56 that stores time series data of an image in which some or all of the masks are released (hereinafter referred to as an unmasked image), and unmasked image data stored in the memory. And a monitor 58.

  In the monitoring system 2 having such a configuration, when an image representing a monitoring area is formed on the imaging surface of the imaging device 18 via the optical system 16 of the imaging unit 4, the imaging device 18 is driven from the control unit 22. An electric signal is output to the A / D converter 20 in synchronization with the signal. The A / D conversion unit 20 performs A / D conversion on the electrical signal and outputs the digital signal to the image memory 23 of the masking unit 6.

  The object extraction unit 24 extracts an object to be masked on the image frame at each time based on the time-series image data stored in the image memory 23. Subsequently, the mask area setting unit 26 sets the position, size, etc. on the image to be actually masked based on the position, size, etc. of the extracted object to be masked on the image, and the setting information data Is output to the mask processing unit 28, and data of mask information (the setting information and the type of object to be masked) is output to the mask information storage unit 30. The masking unit 28 performs mask processing on the time-series image data stored in the image memory 23 to acquire mask image data, and outputs the mask image data to the mask image storage unit 32.

  The mask image data is encoded by the encoding circuit 34, then multiplexed by the multiplexing circuit 36 together with the mask information data, and then sent from the transmission unit 8 to the reception unit 10 via the communication line 38.

  The multiplexed data received by the receiving unit 10 is separated into mask image data and mask information data by the separation circuit 40. The mask image data is further decoded by the decoding circuit 42 and output to the mask image storage unit 44. The mask information data is output to the mask information storage unit 46.

  When a supervisor who does not have a release right (for example, a supervisor whose ID number is S5 in FIG. 8) accesses the system 2 via the input unit 54, the control unit 50 does not output a mask release command to the unmask processing unit 48. The unmask processing unit 48 does not perform mask release processing, and the mask image stored in the mask image storage unit 44 is output to the image memory 56 as it is. The monitor 58 of the display unit 14 displays an image in which all objects are masked as shown in FIG.

  When a monitor who has a release right accesses the system 2 via the input unit 54, the control unit 50 refers to the monitor authority information storage unit 52, and unmasks the information data of the type of the object to be unmasked. Output to 48. The unmask processing unit 48 refers to the mask information stored in the mask information storage unit 46, that is, information such as the type of object hidden by each mask and its position, size, etc. To cancel part or all of the mask. The unmasked image is output to the image memory 56. As a result, the unmasked image is displayed on the monitor 58. For example, when a supervisor (ID number S1) having the authority to unmask only the car accesses the system 2, an image in which only the mask of the car C is unmasked is displayed on the monitor 58 as shown in FIG. As described above, according to the present embodiment, it is possible to change the masks to be released according to the supervisors having different authorities. Therefore, the observer can see an image related to the object he / she needs for monitoring, and can appropriately protect privacy from other objects being masked.

  In this embodiment, the masking method (the method for converting pixels in the region to be masked) is the same regardless of the type of object to be masked, but may be different. In this case, the mask processing unit 28 has a table indicating the correspondence between the object and the mask processing method, and the mask area setting unit 26 includes the object to be masked in addition to the information data of the position and size of the area to be masked. Are output to the mask processing unit 28. Correspondingly, the unmask processing unit 48 also has a table indicating correspondence between objects and mask processing methods. Instead, such a table is stored in the mask information storage unit 30, and the data of the table is transmitted together with the mask information and the image data via the transmission unit 8, and the table data is output to the mask information storage unit 46. In this case, the unmask processing unit 48 does not need a table.

  In this embodiment, the masks processed by the mask processing unit 28 are all reversible. However, when the mask processing method varies depending on the type of object to be masked, an irreversible mask (unmask processing unit 48) is used for some objects. May be applied with a mask that cannot be released. In this case, it is possible to further protect the privacy by irreversibly masking an object that the system administrator or the like always wants to keep secret from the monitor.

  In the above-described embodiment, the imaging unit (camera) 4 is a fixed camera. However, a movable camera that can change the pan angle, the tilt angle, and the zoom in response to a command from the supervisor may be used. In this case, the shape and size of the model used in the object extraction unit 24 change according to the orientation of the camera during the movement of the camera.

  In the present embodiment, the camera 4 is used as the image input unit, and an image obtained by masking a part of the original image captured by the camera 4 is displayed on the monitor 58, so that monitoring can be performed in real time. Instead, a VTR, hard disk, MO (optical disk), CD-R, DVD, or the like may be used. In this case, an image obtained by partially masking the image once stored is displayed on the monitor 58.

Embodiment 2. FIG.
FIG. 11 shows Embodiment 2 of the monitoring system according to the present invention. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted unless necessary. The monitoring system 60 according to the present embodiment is similar to the monitoring system 2 according to the first embodiment, but the transmission unit 8 is configured to transmit the original image together with the mask image. Specifically, the image memory 23 is connected to the multiplexing circuit 36 via the encoding circuit 62. The separation circuit 40 is connected to an image recording unit 64 such as a VTR, hard disk, MO (optical disk), CD-R, or DVD.

  In such a monitoring system 60, based on the image data transmitted from the imaging unit 4 and stored in the image memory 23, a mask process is performed to create a mask image, and the unmasked original image data is encoded. The signal is encoded by the circuit 62 and input to the multiplexing circuit 36. The multiplexing circuit 36 multiplexes the original image data, the mask image data from the encoding circuit 34, and the mask information data from the mask information storage unit 30 to create multiplexed data. On the other hand, the separation circuit 40 separates multiplexed data input via the transmission unit 8, the communication line 38, and the reception unit 10, and mask image data is sent to the decoding circuit 42, and mask information data is sent to the mask information storage unit. 46, the original image data (encoded) is output to the image recording unit 64.

  According to the present embodiment, by displaying a partially masked image and recording the original image, the mask image can be used for monitoring and the original image can be used for storage. This is especially true when some objects are irreversibly masked, and as a result, even if an authorized observer cannot see the original image once masked, The system administrator has an advantage that the stored original image can be viewed as necessary.

Embodiment 3 FIG.
12 and 13 show a monitoring camera and a mask releasing device constituting Embodiment 3 of the monitoring system according to the present invention. Surveillance camera 70 is a camera with a built-in configuration similar to the transmission side of the first embodiment shown in FIG. The mask releasing device 72 is used together with the monitoring camera 70 and has a configuration similar to that of the receiving side of the first embodiment.

  The monitoring camera 70, instead of the mask information storage unit 30, separates the mask information provided from the mask area setting unit 26, and the correspondence between the monitor who monitors the mask image and the types of objects that can be observed by the monitor For example, a mask information / monitorer authority information storage unit 30 ′ for storing a table indicating the correspondence between the ID number of the monitor and the types of objects that can be unmasked as shown in FIG. The data of this table is output to the multiplexing circuit 36 together with the mask information data, and is transmitted via the transmission unit 8 together with the mask image data. On the other hand, the mask release device 72 includes a mask information storage unit 46 ′ that stores the mask information separated from the separation circuit 40 and the table, instead of the mask information storage unit 46. Based on the ID number input from the input unit 54, the control unit 50 refers to the table stored in the mask information storage unit 46 ′ and sends a predetermined signal to the unmask processing unit 48. In this configuration, the supervisor authority information storage unit 52 in FIG. 1 is unnecessary.

  According to the present embodiment, as in the first embodiment, the monitor can see an image related to the object that he / she needs for monitoring, and other objects are masked, so that privacy is appropriately protected. can do. The present embodiment further has an effect that the monitoring system can be made compact because the mask image generation function is built in the monitoring camera.

  The above-described configuration of the present embodiment may be applied to the first embodiment. That is, the mask information storage unit 30 in FIG. 1 stores a table showing the correspondence between the supervisor ID number and the types of objects that can be unmasked as shown in FIG. 8, and the data in the table is mask information and mask image data. If the table is transmitted to the mask information storage unit 46 so that the control unit 50 can refer to the table, the supervisor authority information storage unit 52 is not necessary.

  Also in the present embodiment, similarly to the second embodiment, the surveillance camera may transmit both the mask image and the original image, and the original image may be recorded on the mask release device 72 side.

1 is a block diagram showing Embodiment 1 of a monitoring system according to the present invention. The figure which shows the example of the background image used for the extraction by the object extraction part of FIG. The figure which shows the example of the image of the certain time which the imaging part of FIG. 1 imaged. The figure which shows the difference image of the image of FIG. The figure for demonstrating the model fitting to the moving body extracted in FIG. The figure for demonstrating matching of the mobile bodies between time-sequential images. The figure which shows the example of the mask area | region on the image set by the mask area | region setting part of FIG. The figure which shows the example of the table which shows the correspondence of each monitor's ID number and the object which this monitor can remove a mask memorize | stored in the monitor authority information storage part of FIG. The figure which shows the example of a mask image. The figure which shows the example of the mask image from which the partial mask was cancelled | released. The block diagram which shows Embodiment 2 of the monitoring system which concerns on this invention. The block diagram which shows the surveillance camera used for Embodiment 3 of the surveillance system which concerns on this invention. The block diagram which shows the mask release apparatus used for Embodiment 3 of the monitoring system which concerns on this invention.

Explanation of symbols

2 Monitoring system 4 Imaging unit (image input unit)
6 Masking part 12 Mask releasing part 70 Surveillance camera 72 Mask releasing device

Claims (7)

An image input unit;
When one or more objects to be masked are included in the original image input by the image input unit, a mask image is created by masking the objects, and the position of the mask area for each object on the original image And a masking unit for generating mask information indicating the size and the type of the object,
An unmask processing unit for releasing the mask of the mask image based on the mask information;
A storage unit for storing a table indicating correspondence between a monitor who monitors a mask image and the types of objects which can be observed by the monitor;
When receiving a command from a certain monitor, a control unit that controls the unmask processing unit so as to release only the mask related to the type of object corresponding to the monitor with reference to the table stored in the storage unit;
Monitoring system with.   The monitoring system according to claim 1, wherein the image input unit is a camera.   The monitoring system according to claim 1, wherein the masking unit generates a mask that cannot be released together with a mask that can be released by the unmask processing unit.   The monitoring system according to claim 1, further comprising an image recording unit that records an original image input by the image input unit. An imaging unit;
When one or more objects to be masked are included in the original image captured by the imaging unit, a mask image in which the object is masked is created, and the position of the mask area for each object on the original image and A masking unit for generating mask information indicating the size and the type of the object;
A mask image, mask information, and a transmission unit that transmits data relating to a table indicating correspondence between a monitor that monitors the mask image and the types of objects that the monitor can observe;
With surveillance camera.   The surveillance camera according to claim 5, wherein the transmission unit also transmits data related to the original image captured by the imaging unit. In the mask removal apparatus used with the surveillance camera of Claim 5 or 6,
A receiving unit for receiving data on the mask image, mask information, and table;
An unmask processing unit for releasing the mask of the mask image based on the mask information;
When receiving a command from a certain monitor, based on the above table, a control unit that controls the unmask processing unit so as to release only the mask related to the type of object corresponding to the monitor;
An unmasking device.

Images