KR102012672B1 - Anti-crime system and method using face recognition based people feature recognition - Google Patents

Abstract

The present invention relates to a security system using human feature recognition based on face recognition, and to a method thereof. More specifically, the security system using human feature recognition based on face recognition obtains an image through one or more from a camera configured in a security terminal such as a CCTV installed in a security area such as a road, a building, or the like, an access control device, and the like, and a camera configured in a computer terminal unit such as a laptop, a computer, smart devices, or the like. The security system using human feature recognition based on face recognition determines whether a person included in the obtained image is a person to be recognized with respect to the corresponding security terminal and the computer terminal unit, and performs at least one security responding processing operation among map-based location tracking, access permission or blocking, and use or blocking of the computer terminal in accordance with whether the person is to be recognized.

Description

Anti-crime system and method using face recognition based people feature recognition}

The present invention relates to a security system and method using a face recognition-based human feature recognition, more specifically, a camera, a laptop, a computer configured in a security terminal, such as CCTV, access control devices installed in the security area of the road, building, etc. Acquires an image through at least one of cameras configured in a computer terminal such as a smart device, and determines whether a person included in the acquired image is a recognized person to be recognized for the security terminal and the computer terminal. The present invention relates to a security system and method using face recognition based human feature recognition that performs one or more security response processing operations such as map-based location tracking, access permitting or blocking, using or blocking a computer terminal according to whether a person is recognized.

Video-based security systems such as CCTV systems, DVR systems, access control systems, etc. to protect the public on the streets from various dangers and crimes, to keep the company secure, and to allow only those authorized to enter the security area. Are being installed on streets and buildings.

In general, security and surveillance security systems such as CCTV, DVR, etc. are required for 24 hours surveillance, security vulnerable areas around residential areas, schools, playgrounds, vulnerable points that employees cannot always manage, banks, department stores, airports, ports, companies By installing equipment such as cameras in facilities such as facilities, and transmitting and displaying and recording field images in real time to a central control center connected to the network, the situation manager monitors various accidents, dangerous situations and crimes in advance. Preventive, safety checks and various information can be taken to take follow-up measures or remedial measures, and less manpower can be used to identify abnormalities in many monitoring sites and provide essential information for future criminal investigations and finding missing persons. do.

In addition, when the building is provided with areas that allow access according to the security level is provided with a gate type access control device installed in the entrance and the wall type access control device installed in the door. Such an access control device captures an image of a person who wants to enter a security area through a gate and a door, and photographs a face part when a face authentication request is performed, and performs accessor authentication according to face recognition to allow or deny access according to authentication success. Perform a block.

However, conventional security systems such as CCTV, DVR, etc. do not perform face recognition by real-time analysis of collected and developed images with the focus on acquiring, displaying and storing images, and do face recognition by analyzing real-time or stored images. Even if it is performed, there is a problem that face recognition is impossible when the user in the image covers the face with a mask, glasses, a hat, or the like, or turns the face in a direction other than the camera direction.

In addition, security systems, such as access control devices, have been developed with the focus on identifying the subjects through real-time face recognition, but even in this case, if the user covers the face with a mask, glasses, hat, etc., or turns the face, the face recognition is performed. There was this impossible problem. Therefore, the security system such as the access control device induces the recognition subject to look at the camera in front, but the side of the recognition subject has been inconvenient because each time the entrance and exit, only to take off the glasses, hats, masks, etc. to look at the camera.

In order to improve this problem, the Republic of Korea Patent No. 10-1215948 [image information masking method of the surveillance system based on the body information and facial recognition] registered the body size and color information of the clothing to be recognized For example, the present invention discloses a technique for identifying a person to be recognized based on body size and clothing birth information when face is not recognized.

However, there is a problem that the body size can be changed according to the standing position of the subject, the height of the shoe, etc., and the recognition rate is extremely low. The clothing is also frequently changed, which is useful only for searching for missing children, and is applicable to general recognition of the subject. There was a problem that could not be.

In addition, there is a problem that the technique using the body size is not applicable to the access control device mainly photographing the upper body.

Patent Registration No. 10-1215948 (As of December 27, 2012) Patent Registration No. 10-1416568 (2014.07.10.)

Therefore, an object of the present invention is an image through any one or more of the camera configured in the security terminal, such as CCTV, access control devices installed in the security area, such as roads, buildings, and the camera terminal unit, such as laptops, computers, smart devices, etc. Determine whether the person included in the acquired image is the person to be recognized for the security terminal and the computer terminal unit, and map based location tracking, access permitting or blocking, computer A security system and method using face recognition based human feature recognition to perform one or more security response processing operations such as terminal use or blocking.

Security system using a face recognition-based human feature recognition according to the present invention for achieving the above object: at least one or more to obtain an image of the security area to transmit the device identification information and the image information including the image; Image information acquisition unit; And when the image information is received, extract the device identification information and the image, and perform at least one of face recognition and human appearance feature recognition from the extracted image to perform recognition of the recognition target, whether to recognize the recognition target, the recognition target and The image information acquisition unit performs a corresponding security process corresponding to the installed security area, and upon identification of a person to be recognized successfully, person appearance characteristic information generated by the person appearance feature recognition is mapped to the recognized person to be recognized, and And a security manager to identify the person to be recognized by comparing the person appearance feature information acquired from the image when the face recognition fails with the person appearance feature information registered when the face recognition is successful.

The security management unit, the image storage unit for receiving and storing the image information; And when the image information is received, extract the device identification information and the image, and perform at least one of face recognition and human appearance feature recognition from the extracted image to perform recognition of the recognition target, whether to recognize the recognition target, the recognition target and The image information acquisition unit performs a corresponding security process corresponding to the installed security area, and upon identification of a person to be recognized successfully, person appearance characteristic information generated by the person appearance feature recognition is mapped to the recognized person to be recognized, and And a security processor for identifying the person to be recognized by comparing the person appearance feature information acquired from the image when the face recognition fails with the person appearance feature information registered when the face recognition is successful.

The security management unit includes a device DB for storing device information including the image information DB for receiving and storing the image information, the device identification information, and the device location information, and accessing, accessing, and using the device in the security area for each security area. A storage unit including a subject DB for storing personal information, facial feature information, and human appearance characteristic information about the subject who defines any one of permission and blocking of one or more; A communication unit configured to access a wired / wireless data communication network and perform data communication with image information acquisition units connected to the wired / wireless data communication network; Receiving image information from at least one image information acquisition unit through the communication unit, extracting device identification information and image, and performing at least one of face recognition and human appearance feature recognition from the extracted image to identify a person to be recognized. And perform a corresponding security process corresponding to whether or not to recognize a recognized person, a recognized person, and a security area in which the image information acquisition unit is installed, and to recognize the recognized person's feature information generated by the person's appearance feature recognition. And a security processing unit including a control unit which maps and registers and compares the external appearance feature information acquired from the image when the face recognition fails with the external appearance feature information registered when the face recognition succeeds. It is done.

When the face registration request is generated from the image information acquisition unit, the controller acquires administrator authentication information through the image information acquisition unit, performs administrator authentication, and when the administrator authentication is successful, an image inputted to the recognition target through the image information acquisition unit. A face registration unit which extracts facial feature information from an image of information and maps the feature information to personal information of the subject to be registered in the subject DB; A security event occurrence monitoring unit for monitoring whether a security event by an administrator occurs from the image information acquisition unit or with respect to the image information acquisition unit through a communication unit; An image information collecting unit collecting image information from at least one image information obtaining unit in which the crime event has occurred; Equipment type determination unit and the equipment to extract the device identification information from the image information collected through the image information collecting unit, and to determine the type of equipment by loading information on the corresponding equipment from the equipment DB by the equipment identification information An equipment information acquisition unit including an equipment location determination unit to determine a location of equipment based on identification information; A person recognition processor that recognizes a person to be recognized by performing face recognition from the image of the image information collected through the image information collecting unit, and performs recognition of a person's appearance by recognizing a person's external features when the recognition of the person to be recognized by the face recognition fails. ; And a security response processor configured to perform a security response processing operation according to the type and equipment location of the corresponding security area when the recognition target person is identified in the person recognition processing unit.

The storage unit further includes a map DB for storing map data including the location of the image information acquisition unit and a security area, and the control unit is further configured to determine the location tracking target according to the type of the device and whether the target is identified. The location of the location tracking target is tracked by the image received from the image information acquisition unit for the location tracking target and the location of the image information acquisition unit where the corresponding image is received, and the displayed location is matched with the map data and displayed or stored. Characterized in that it further comprises a location tracking unit.

The storage unit further includes an administrator DB for storing administrator information including administrator personal information for each security area and notification means information of the administrator. Security information for notifying the administrator in charge of the security area with reference to the administrator DB, the security information including the detection target information of the illegal recognition target and the location information of the image information acquisition unit accessed by the illegal recognition target. It further comprises a notification unit.

The person recognition processing unit may include a person detection unit that detects and outputs a human image from the image; A face detection unit detecting a face in the human image and detecting face feature information according to the appearance of the face; A face recognition unit comparing the face feature information detected by the face detector with the face feature information of the DB of the person to identify the person to be recognized; A human feature extracting unit extracting human external feature information from the human image if the object to be recognized through the face recognition unit cannot be identified; And a feature recognizing unit for identifying the person to be recognized by comparing the extracted person person feature information with the person person feature information registered in advance in the person to be recognized DB.

The image information acquisition unit is a security camera CCTV camera, the security response processing unit of the control unit, characterized in that to track the location of the recognition target by driving the location tracking unit when the recognition target is identified.

The image information acquisition unit may be a computer and a smart device, wherein the security event occurrence monitoring unit monitors whether user authentication request information including computer identification information, which is a security event, is received from the computer, and the image information collection unit is equipment identification information. Monitor whether the video information including the computer identification information is collected from the smart device, and output the information to the device information acquisition unit and the person recognition processing unit. When the recognition target is identified, the computer notifies the authentication success information to allow the computer to be used, and when the recognition target is not identified, the authentication failure information is transmitted to the computer to block the use of the computer. .

Security method using the face recognition-based human feature recognition according to the present invention for achieving the above object: the control unit of the security processing unit collects the image information from the at least one image information acquisition unit through the image information collection unit each image information Image information collection process of extracting the equipment identification information and the image of the; The control unit detects the facial feature information and the human external feature information from the image extracted by the human recognition processor, and recognizes the subject by any one or more of facial recognition by the facial feature information and human external feature recognition by the human feature information. Identifying the identifying subject; Registering the person appearance feature information by mapping the person appearance feature information generated by the person appearance feature recognition to the recognized person when the identification of the person to be recognized is successful; And a person appearance-based recognition object identification process for identifying the recognition person by comparing the human appearance feature information obtained from the image when the recognition object identification failure by the face recognition and the person appearance feature information registered when the face recognition is successful. It is characterized by.

The method may include: when the face registration request is generated from the image information acquisition unit through the face registration unit, the controller obtains administrator authentication information through the image information acquisition unit, and performs administrator authentication to perform the administrator authentication. The method may further include a face registration process of extracting facial feature information from an image of image information input to a person to be recognized, mapping it to personal information of the person to be recognized, and registering it in the person to be recognized DB.

The method may include: an apparatus in which the controller extracts device identification information from the image information collected through the image information collecting unit, and loads information on a corresponding device in a device DB based on the device identification information to determine the type of equipment. A device information acquisition process comprising a device location determination step of determining a location of the device by a type determination step and the device identification information; And a security response processing process of performing, by the controller, a security response processing operation according to the type and equipment location of the corresponding security area to the recognition target recognized by the security response processing unit when the recognition target person is identified in the person recognition processing unit. Characterized in that.

The method may include: when the control unit determines a location tracking target based on the type of equipment and whether or not the target is identified through the location tracking unit, an image received from the video information acquisition unit for the location tracking target and an image information acquisition unit where the corresponding video is received. The method may further include a location tracking process of tracking the location of the location tracking subject by location and displaying or storing the tracked location after matching the map data.

The method may include: acquiring recognition object detection information, which is information of an illegal recognition object, and image information accessed by an illegal recognition object if the recognition object according to the device type and the device type is an illegal recognition object through the security information notification unit. And a security information notification process of notifying the administrator in charge of the security area by referring to the administrator DB for security information including negative location information.

The recognition object identification process may include: a person detection step of the control unit detecting and outputting a human image from the image through a person detection unit; A face detection step of the control unit detecting a face in the human image through a face detection unit and detecting face feature information according to the appearance of the face; A face recognition step of the control unit identifying a recognition target by comparing the face feature information detected by the face detection unit with the face feature information of the recognition target DB through a face recognition unit; Extracting human feature information from the human image when the controller fails to identify the person to be recognized through the face recognition unit through the human feature extractor; And a feature recognizing step of identifying, by the control unit, a recognized person by comparing the extracted human appearance feature information with the extracted person appearance feature information previously registered in the recognition person DB.

If the image information acquisition unit is a CCTV as a security terminal, the control unit, if the recognition target is identified in the security response processing, performs the location tracking process in a security response processing to track the location of the recognition target. do.

If the image information acquisition unit is a computer and a smart device. In the security response process, when the person to be recognized is identified through the person recognition processing unit, the control unit notifies the computer of authentication success information to allow the computer to be used, and if the person to be recognized is not identified, the computer fails to authenticate. And transmits information to perform a security response processing operation that blocks the use of the computer.

The present invention has the effect of identifying at least one or more recognition subjects from an image obtained through a camera of the CCTV and DVR system by applying to a security system such as CCTV and DVR system.

In addition, the present invention can extract the external features of the subject as well as the face to identify the subject by the external features in a situation where the face is not recognized, it can provide convenience to the subject, the appearance features Since the location tracking is performed until the recognition target is successfully recognized by the face recognition, the security target can be maintained.

In addition, the present invention extracts and registers a pattern and a color pattern of a garment every time the face recognition is successful in a certain period of time (for example, one day) among the external features, thereby increasing the recognition rate of the subject to be recognized by the garment of the recognized subject. Has

In addition, the present invention allows the use of the computer terminal based on the facial recognition of the image obtained through the camera of the PC camera, notebook camera and smart devices as well as the security system for the security terminal, etc. It has the effect of improving security.

In addition, the present invention performs the location tracking of the recognized subject on the blacklist using the map, the subject of the missing reporter or the person's appearance features, so the recognition of the subject of the blacklist, missing or recognized recognition of the temporary list It has an effect that can quickly cope with the occurrence of the event by the recognized object.

1 is a view showing the configuration of a security system using a face recognition-based human feature recognition in accordance with the present invention.
2 is a view showing the detailed configuration of the security processing unit of the security system according to the present invention.
3 is a diagram illustrating a map data and equipment information mapping screen in a security system using face recognition-based human feature recognition according to an embodiment of the present invention.
4 is a flowchart illustrating a security method using face recognition based human feature recognition in the security processing unit of the security system according to the present invention.
5 is a flowchart illustrating a security method using face recognition based human feature recognition when a security event occurs from a security terminal according to an embodiment of the present invention.
6 is a flowchart illustrating a security method using face recognition based human feature recognition when a security event occurs from a computer terminal according to an embodiment of the present invention.
7 is a flowchart illustrating a face recognition method among security methods using face feature-based human feature recognition according to the present invention.

Hereinafter, a configuration and operation of a security system using face recognition based human feature recognition according to the present invention will be described with reference to the accompanying drawings, and a security method using face recognition based person feature recognition and guidance in the system will be described in detail. do.

1 is a view showing the configuration of a security system using a face recognition-based human feature recognition in accordance with the present invention.

Referring to FIG. 1, a security system using face recognition based human feature recognition according to the present invention includes at least one image acquisition unit 100 and a security management unit 200.

The image acquisition unit 100 and the security management unit 200 are connected through the wired or wireless data communication network 10 to perform mutual data communication.

The wired / wireless data communication network 10 includes an Internet network including a wide area network (WAN) including a Wi-Fi network and a local area network (LAN), a third generation (3 Generation), and 4G. , At least one of a mobile communication network including 5G, a WiBro network, a dedicated network, and the like are combined.

The image information acquisition unit 100 is a device installed in an arbitrary security area and includes a camera for capturing the security area. The image information acquisition unit 100 includes image information including an image photographing the security area and equipment identification information which is its own unique identification information. It generates and transmits to the security management unit 200 through the wired and wireless data communication network (10).

As shown in FIG. 1, the image information acquisition unit 100 may include one or more computer terminals, such as a personal computer, a computer 110 such as a notebook computer, a smart device 120 such as a smart phone, a smart pad, etc. It may be configured as a computer terminal unit including, may be an access control device (130, 140) which is a security terminal for controlling the access of the security area, the camera 150 of the CCTV system and NVR system as a security terminal (described below) May be referred to as a "CCTV camera" for convenience.

The wall type access control device 130, the gate type access control device 140, and the CCTV camera 150, which are security terminals, are all composed of a camera module 1, and are images captured by the camera module 1, respectively. Image information including the device identification information of the transmission through the wired and wireless data communication network 10 to the security management unit 200.

The security manager 200 collects image information from the image information acquisition unit 100, extracts equipment identification information and images of the collected image information, recognizes face recognition and human features from the image, and recognizes face recognition and human features. It is determined whether or not the person recognized by any one of them is a recognized person, and a security response processing operation is performed according to whether the person recognized is a recognized person. The security response processing may be operations such as tracking the location of the recognized person, blocking access to the security area, allowing access, temporary approval, allowing the use of the computer terminal, and blocking the use of the computer terminal.

The security management unit 200 may be configured only with the security processing unit 400 that performs the overall operation, and further includes an image storage unit 300 that stores only the image information and provides the corresponding image information at the request of the security processing unit 400. It may be configured to include.

The security management unit 200 first performs face recognition of a person in the security area, and recognizes whether at least one or more people present in the security area by the face recognition are suitable or not suitable for the security area. If the person is wearing a hat, mask, glasses, etc., or turns his face and does not recognize the face, it is identified by the feature recognition of the person. The above features may include types of clothes such as height, short sleeves, long sleeves, skirts, pants, dresses, blouses, suits, jumpers, hoods, clothing color patterns, short hairs, hair styles, hairstyles such as hair lengths, hair colors Could be.

2 is a view showing a detailed configuration of the security processing unit of the security management unit of the security system according to the present invention, Figure 3 is a map data and the security system using a face recognition-based human feature recognition in accordance with an embodiment of the present invention; Fig. Shows the equipment information mapping screen. A description with reference to FIGS. 2 and 3 is as follows.

The security processor 400 includes a storage 410, a communication unit 420, and a control unit 430.

The storage unit 410 includes a device DB 411, a map DB 412, a target DB 413, a manager DB 414, an image information DB 415, and a location tracking DB 416.

The equipment DB 411 stores and manages equipment information about the security terminals 130, 140, and 150, the computer 110, and the smart devices 120 of the image information acquisition unit 100. As shown in the table of FIG. 3, the device information includes a device identification number, a device name (name), including any one or more of a unique identifier (ID), a MAC address, and a phone number for the devices. The location information of the security area where the equipment is installed may be included. The location information may include place identification information, building and floor information of a place, coordinate information on map data, and the like.

Map DB 412 stores the map data for the security area as shown in FIG.

The recognition target DB 413 stores personal information about the recognition target to be recognized in the security area, facial feature information for face recognition, and human appearance feature information. The person to be recognized may be a lost person, a missing person, an accessor, a blacklisted person, or the like.

The manager DB 414 stores manager information for each security zone. The manager information may include a manager's phone number, an ID such as a messenger, manager notification means information such as an e-mail, manager authentication information, and manager personal information.

The image information DB 415 stores image information including equipment identification information and images.

The location tracking DB 416 stores location tracking information for each location tracking object.

The communication unit 420 connects to the wired / wireless data communication network 10 and performs data communication with other devices connected to the wired / wireless data communication network 10.

The controller 430 controls the face register 440, the crime event occurrence monitoring unit 450, the image information collecting unit 460, the equipment information obtaining unit 470, the person recognition processing unit 480, and the security response processing unit 490. It further includes a location tracking unit 500 and security information notification unit 510 according to the embodiment to control the overall operation of the security processing unit 400 according to the present invention.

The face register 440 obtains administrator authentication information through the image information acquirer 100 when a face registration request is generated from the image information acquirer 100, and is stored in the obtained manager authentication information and the manager DB 414. Comparing the manager authentication information to determine whether the authentication is successful, and after the authentication is successful, the facial feature information is extracted through the person recognition processing unit 480 to be described later, the image of the video information input through the image information acquisition unit 100, It is stored in the subject DB 413 and registered.

The security event occurrence monitoring unit 450 monitors whether a security event occurs from the image information acquisition unit 100. The security event may be generated when the driving of the image information acquisition unit 100 starts, or may be generated when an authentication request based on face recognition is received from the image information acquisition unit 100. It may be generated upon a start request. In addition, the security event occurs when the image information acquisition unit 100 is composed of a computer 110 and a smart device 120, when the smart device user receives the user authentication request information transmitted when the user wants to use the computer terminal Could be

The image information collecting unit 460 collects image information from the driven image information obtaining unit 100 and stores the image information in the image information DB 415. When the security storage unit 300 is provided, the image information is performed by the security storage unit 300 and stored and managed in the internal DB of the security storage unit 300, the image information collection unit 460 is the It may be received through the secure storage 300. In another embodiment, the image information collecting unit 460 may be configured to receive and process the image information when necessary to collect the image information from the image information acquisition unit 100 and store the image information in the image storage unit 300. will be.

In addition, the image information collection unit 460 is a computer that is equipment identification information of the computer 110 from the smart device 120, when the image information acquisition unit 100 is composed of a computer 110 and a smart device 120 Receives and processes video information including identification information.

The device information acquisition unit 470 extracts device identification information from the image information collected by the image information collecting unit 460, and determines the type of the device. It includes a device position determination unit 472 for determining information.

The person recognition processor 480 may include a person detector 481, a face detector 482, a face recognizer 383, a person feature extractor 484, and a feature recognizer 485. Face recognition based recognition subject identification is performed according to the determined equipment type and equipment location.

Specifically, the person detector 481 determines whether a person exists in the image of the image information received from the image information acquisition unit 100 of the acquired equipment information when the equipment information is acquired by the equipment information acquisition unit 470. If present, a human image is extracted from the image.

The face detector 482 detects a face from a human image when a person is detected through the person detector 481, and extracts facial feature information from the detected face image.

The face recognition unit 483 compares the face feature information extracted by the face detector 482 with the face feature information registered in the person DB 413 to identify a person to be recognized for matching face feature information.

When a person is detected by the person detector 481, the person feature extractor 484 extracts person appearance feature information of the person from the detected person image. The human appearance characteristic information may include a height, a type of clothes, a color pattern of clothes, a short hair, a hair shape, a hair style, a hair color, and the like.

The human feature extractor 484 maps the detected human appearance feature information to the corresponding person information of the person DB 413 whenever the person to be recognized is identified by successful face recognition through the face recognition unit 483. Save it.

The feature recognizing unit 485 is stored in the human appearance feature information extracted from the human feature extracting unit 484 and stored in the recognition target DB 413 when the recognition target based on the face recognition is not identified through the face recognition unit 483. By comparing the human feature information, a person to be recognized having the same person feature information is identified to identify a person to be recognized by the person feature information.

The security response processing unit 490 performs a security response processing operation according to the type and location of the equipment recognized by the device information acquisition unit 470 to the recognition target recognized by the person recognition processing unit 480. The security response processing operation may include door opening, door blocking, alarm generation, image storage, administrator notification, etc. when the equipment, that is, the image information acquisition unit 100 is a gate type and a wall type access control device, is a CCTV camera. In this case, it may be an alarm occurrence, location tracking, image storage, notification of the subject information manager to the recognized person, and in the case of a computer terminal, it may be allowed to use a computer, set a computer lock, and activate a protection screen.

When the image information acquisition unit 100 is a computer 110 and a smart device 120 that do not have a PC camera, the security response processing unit 490 may be activated when the computer 110 is logged in or in a standby state. When generated, when the user authentication request information including the computer identification information from the computer 110 is received from any smart device 120 includes an image including the computer identification information and the user photographed through the smart device 120 Receives the image information to recognize the recognition target through the face recognition and human feature recognition through the image, and if the recognition target for the security area in which the computer 110 is installed is a legitimate recognition target authentication information to the computer 110 Notify and allow the user to use the computer 110.

If the location tracking target is determined according to the type of the equipment and whether the recognition target is identified, the location tracking unit 500 may receive an image received from the video information acquisition unit 100 and the video information acquisition unit that receives the corresponding video. The location of the location tracking target is tracked by the location of the location 100, the tracked location is matched with the map data, and then displayed on a display device (not shown) and a manager terminal (not shown) or the storage unit 410. Store in the location tracking DB (416).

For example, if the position tracking unit 500 does not recognize the face and only the person feature recognition by the person appearance feature information is successful, the location tracking unit 500 may perform the face recognition based on the person feature recognition until the face recognition is performed on the person identified by the person feature recognition. The location tracking of the subject is performed.

In addition, the location tracking unit 500 performs location tracking based on face recognition and human feature recognition when the recognition target is a recognition target set as a tracking target.

The security information notification unit 510 may recognize recognition object detection information, which is information on the recognized recognition object, and the image information acquisition unit 100 that detects the recognition object when the administrator notification is determined among the security response methods taken by the security response processing unit 490. Location), that is, security information including the equipment location information, is provided to the manager corresponding to the security area in which the image information acquisition unit 100 in which the recognition target is photographed is referred to the manager DB 414.

The security information notification unit 510 may be provided to the administrator's mobile phone, smartphone, etc. through the mobile communication message of the mobile communication technology such as 2G (2G), 3G, 4G, 5G, etc. It may be provided through a push message of an application, may be provided through a message through a messenger, or may be provided through an email.

4 is a flowchart illustrating a security method using face recognition based human feature recognition in the security processing unit of the security system according to the present invention.

Referring to FIG. 4, the controller 430 monitors whether a security event occurs from any image information acquisition unit 100 (S111). When the crime event occurs, the image information acquisition unit 100 that has generated the crime event begins to acquire the image information (S113).

When the image information starts to be acquired, the controller 430 recognizes the device, that is, the image information providing unit 100, and the location thereof by the device identification information included in the obtained image information (S114).

In addition, when the device is recognized, the controller 430 recognizes the face of the person included in the acquired image information (S115), and the person recognized by the face recognition is registered in the subject DB 413. In operation S117, it is determined whether the image information acquisition unit 100 provided with the recognition target is the installed security area.

If it is determined that the identified person is the person to be recognized, the controller 430 stores an image and starts a security response processing operation corresponding to the person to be recognized for the security area (S119), and determines that the identified person is not the person to be recognized. When the image is stored (S121), the security response process for the unrecognized person is started (S123).

After the start of the security response processing operation, the control unit 430 sends the security notification information according to the security response processing operation to the manager (S125). If the image information acquisition unit 100 is a CCTV camera, the security response processing operation may not be started for the non-person.

5 is a flowchart illustrating a security method using face recognition based human feature recognition when a security event occurs from a security terminal according to an embodiment of the present invention.

Referring to FIG. 5, the controller 430 checks whether a security event occurs from any security terminal (S211).

When the crime event occurs, the control unit 430 starts to acquire the image information from any security terminal (S213), and identifies the equipment by the equipment identification information included in the acquired image information and identifies the location (S214) .

When the equipment is identified and its location is identified, the controller 430 recognizes a human face of the image information (S215), and determines whether the person identified by the face recognition is a person to be recognized (S217).

If not the recognition target, the control unit 530 determines whether the equipment recognized through the equipment recognition is CCTV (S223), the access control device (S225).

If it is CCTV, the control unit 430 stores the image information in the image information DB 415 of the storage unit 410 (S233).

On the other hand, if the recognized equipment is an access control device, the control unit 430 blocks the entrance by keeping the door of the security area in which the access control device is installed in a closed state (S235), based on the recognized face recognition of the person. In step S237, the location tracking is started until it leaves the security area or the predetermined specific area.

And if it is determined that the recognition target by the face recognition, the control unit 430 determines whether the equipment obtained the image information is the access control device (S219) CCTV camera (S221).

If it is determined that the equipment is CCTV, the control unit 430 stores the image information in the image information DB 415 of the storage unit, initiates the location tracking (S229), and tracks the location information and the recognition target information on the recognized person. Security notification information that is included is sent to the administrator of the security area (S231).

On the other hand, if the equipment is an access control device, the control unit 430 stores the image information and allows access by opening the door and the gate in which the access control device is installed (S227).

6 is a flowchart illustrating a security method using face recognition based human feature recognition when a security event occurs from a computer terminal according to an embodiment of the present invention.

Referring to FIG. 6, the controller 430 monitors whether a security event occurs by receiving user authentication request information including computer identification information from the computer 110 (S311).

When a security event is generated by the computer 110, the controller 430 extracts computer identification information from the user authentication request information (S313).

When the computer identification information is extracted, the controller 430 checks whether image information including the computer identification information and an image is received from an arbitrary smart device 120 (S315).

When the image information is received, the controller 430 performs equipment recognition and equipment position recognition (S316), and performs face recognition (S317).

When face recognition is performed, the controller 430 determines whether the person identified by the face recognition is a person registered in the person DB 413 (S319). It transmits to the computer 110 having (S321), and if not the recognition target computer security setting request information is transmitted to the computer (110) (S323).

The computer receiving the computer use approval information is activated to provide a function that the user wants to use.

On the other hand, the computer that receives the computer security setting request information may set a system lock or a screen saver to prevent access to the main screen so that the computer cannot be used. The system lock setting may be an operation of preventing a system from responding to an input although the main screen is displayed, or an operation of turning off the system.

7 is a flowchart illustrating a face recognition method among security methods using face feature-based human feature recognition according to the present invention.

Referring to FIG. 7, the controller 430 checks whether a person is included in the image, and extracts a person image if the person is included (S411).

When the human image is extracted, the controller 430 extracts a facial image from the extracted human image (S413), extracts facial feature information by the extracted facial image, and extracts the facial feature information from the extracted facial image. Face recognition comparing the facial feature information is performed (S415), and it is determined whether the recognition of the recognition target is successful according to the face recognition success (S417).

If the face recognition is successful, the controller 430 stores and outputs the recognized recognition person's information and location information (S419).

In addition, if the face recognition is successful, the controller 430 extracts the human appearance feature information from the extracted human image, and registers the extracted human appearance feature information by mapping the recognition person information identified by the face recognition (S421). ).

On the other hand, if face recognition fails, the control unit 430 extracts the human external feature information from the human image (S423), and compares the extracted human external feature information with the human external feature information registered in the recognition target DB 413. The recognition of the subject is performed (S425).

After performing the recognition of the person, that is, the recognition target person, according to the external appearance characteristic information, the controller 430 determines whether the person recognition is successful (S427), and outputs information that cannot be recognized when the recognition fails (S429), and when the recognition is successful, As described above, recognition information is output (S431).

On the other hand, the present invention is not limited to the above-described typical preferred embodiment, but can be carried out in various ways without departing from the gist of the present invention, various modifications, alterations, substitutions or additions in the art Anyone who has this can easily understand it. If the implementation by such improvement, change, replacement or addition falls within the scope of the appended claims, the technical idea should also be regarded as belonging to the present invention.

[Description of the code]

100: image information acquisition unit 110: computer

120: smart device 200: security management unit

300: image storage unit 400: security processing unit

410: storage unit 411: equipment DB

412: map DB 413: subject DB

414: manager DB 415: video information DB

416: location tracking DB 420: communication unit

430: control unit 440: face register

450: crime event occurrence monitoring unit

460: image information collecting unit

470: equipment information acquisition unit 471: equipment type determination unit

472: Equipment position determination unit 480: Person recognition processing unit

481: person detector 482: face detector

483: Face recognition unit 484: Person feature extraction unit

485: feature recognition unit 490: security response processing unit

500: location tracking unit 510: security information notification unit

Claims (17)

At least one image information acquisition unit for acquiring an image photographing a security area and transmitting equipment identification information and image information including the image; And
When the image information is received, the device identification information and the image is extracted, and the recognition object identification is performed by performing any one or more of facial recognition and human appearance feature recognition from the extracted image, whether to recognize the recognition object, the recognition target and the The image information acquiring unit performs a corresponding security process corresponding to the installed security area, and when the identification of the object to be recognized succeeds, by mapping the human appearance characteristic information generated by the human appearance characteristic recognition to the recognized recognition object and registering the face. Comprising a security management unit for identifying the recognition target by comparing the human appearance feature information obtained from the image when the recognition failure and the person appearance feature information registered when the face recognition success,
The security management unit,
A device DB for storing device information including the image information DB for receiving and storing the image information, the device identification information, and the device location information, and allowing one or more of access to, access to, and use of the device by security area. And a storage unit including a recognition DB for storing personal information, facial feature information, and human appearance characteristic information about the recognition target defining any one of the blocking.
A communication unit configured to access a wired / wireless data communication network and perform data communication with image information acquisition units connected to the wired / wireless data communication network; And
Receiving identification information by receiving image information from at least one image information obtaining unit through the communication unit, extracting equipment identification information and image, and performing any one or more of facial recognition and human appearance feature recognition from the extracted image. And perform a corresponding security process corresponding to whether a recognized person is identified, a recognized person, and a security area in which the image information acquisition unit is installed, and map the person's appearance feature information generated by the person's appearance feature recognition to the recognized person. And a security processing unit including a control unit for identifying the recognition person by comparing the human appearance feature information acquired from the image when the face recognition fails with the registered person appearance feature information when the face recognition fails.
The control unit,
When a face registration request is generated from the image information acquisition unit, the administrator authentication information is acquired through the image information acquisition unit and the administrator authentication is successful. From the image of the image information input to the recognition target through the image information acquisition unit A face register extracting face feature information and mapping the face feature information to personal information of the person to be recognized in the person DB;
A security event occurrence monitoring unit for monitoring whether a security event by an administrator occurs from the image information acquisition unit or with respect to the image information acquisition unit through a communication unit;
An image information collecting unit collecting image information from at least one image information obtaining unit in which the crime event has occurred;
Equipment type determination unit and the equipment to extract the device identification information from the image information collected through the image information collecting unit, and to determine the type of equipment by loading information on the corresponding equipment from the equipment DB by the equipment identification information An equipment information acquisition unit including an equipment location determination unit to determine a location of equipment based on identification information;
A person recognition processor that recognizes a person to be recognized by performing face recognition from the image of the image information collected through the image information collecting unit, and performs recognition of a person's appearance by recognizing a person's external features when the recognition of the person to be recognized by the face recognition fails. ; And
When the recognition target person is identified in the person recognition processing unit, a face recognition-based person feature comprising a security response processing unit for performing a security response processing operation according to the type and equipment location of the security area for the recognized recognition target person. Security system using recognition.
delete delete delete The method of claim 1,
The storage unit,
Further comprising a map DB for storing the map data including the location of the image information acquisition unit and the security area,
The control unit,
When the location tracking target person is determined according to the type of equipment and whether to recognize the recognition target, the location of the location tracking target is determined by the image received from the image information acquisition unit and the location of the image information acquisition unit where the corresponding image is received. Security system using a face recognition-based human feature recognition, characterized in that it further comprises a location tracking unit for tracking and displaying or storing the tracked location to match the map data.
The method of claim 1,
The storage unit,
Further comprising an administrator DB for storing administrator information including administrator personal information for each security area and notification means information of the administrator,
The control unit,
If the recognition target according to the type of equipment and the type of equipment is an illegal recognition target, the administrator DB includes security information including detection information detection information, which is information of illegal recognition target, and location information of the image information acquisition unit accessed by the illegal recognition target. Security system using a face recognition-based person feature recognition, characterized in that it further comprises a security information notification unit for notifying the administrator in charge of the security area with reference to.
The method of claim 1,
The person recognition processing unit,
A person detector for detecting and outputting a human image from the image;
A face detection unit detecting a face in the human image and detecting face feature information according to the appearance of the face;
A face recognition unit comparing the face feature information detected by the face detector with the face feature information of the DB of the person to identify the person to be recognized;
A human feature extracting unit extracting human external feature information from the human image if the object to be recognized through the face recognition unit cannot be identified; And
And a feature recognizing unit for identifying a person to be recognized by comparing the extracted person's external feature information and the person's external feature information pre-registered in the person to be recognized DB.
The method of claim 5,
The image information acquisition unit,
CCTV camera as a security terminal,
The security response processing unit of the control unit,
Security system using a face recognition-based human feature recognition, characterized in that for tracking the location of the recognition target by driving the location tracking unit when the recognition target is identified.
The method of claim 1,
The image information acquisition unit,
Computers and smart devices,
The security event occurrence monitoring unit,
Monitor whether user authentication request information including computer identification information, which is a security event, is received from the computer;
Video information collecting unit,
Monitors whether image information including the computer identification information, which is equipment identification information, is collected from the smart device, and outputs the collected information to the equipment information acquisition unit and the person recognition processing unit,
The security response processing unit of the control unit,
When the person to be recognized is identified through the person recognition processing unit, the computer notifies the user of the authentication success information to allow the computer to be used, and if the person to be recognized is not identified, the authentication failure information is transmitted to the computer to use the computer. Security system using face recognition-based human feature recognition, characterized in that blocking.
A video information collection process of the control unit of the security processing unit to collect the video information from the at least one video information acquisition unit through the video information collection unit to extract equipment identification information and video of each video information;
When the face registration request is generated from the image information acquisition unit through the face registration unit, the controller acquires administrator authentication information through the image information acquisition unit, and performs administrator authentication. A face registration process of extracting facial feature information from the image of the image information input for the mapping and mapping the face feature information to the personal information of the recognition subject and registering it in the recognition subject DB;
The control unit detects the facial feature information and the human external feature information from the image extracted by the human recognition processor, and recognizes the subject by any one or more of facial recognition by the facial feature information and human external feature recognition by the human feature information. Recognizing subject identification process to identify the;
Registering the person appearance feature information by mapping the person appearance feature information generated by the person appearance feature recognition to the recognized person when the identification of the person to be recognized is successful; And
And a person appearance-based recognition object identification process for identifying the recognition person by comparing the human appearance feature information obtained from the image when the recognition object identification failure by the face recognition and the person appearance feature information registered when the face recognition is successful. Security method using facial feature based human feature recognition.
delete A video information collection process of the control unit of the security processing unit to collect the video information from the at least one video information acquisition unit through the video information collection unit to extract equipment identification information and video of each video information;
The control unit detects the facial feature information and the human external feature information from the image extracted by the human recognition processor, and recognizes the subject by any one or more of facial recognition by the facial feature information and human external feature recognition by the human feature information. Recognizing subject identification process to identify the;
Registering the person appearance feature information by mapping the person appearance feature information generated by the person appearance feature recognition to the recognized person when the identification of the person to be recognized is successful;
A person appearance-based recognition object identification process of identifying the recognition person by comparing the human appearance feature information obtained from the image when the recognition object identification failure by the face recognition and the person appearance feature information registered when the face recognition is successful;
A device type determination step of the controller extracting device identification information from the image information collected through the image information collecting unit, and determining the type of the device by loading information on the corresponding device in the device DB based on the device identification information; A device information acquisition process comprising a device location determination step of determining a location of the device by the device identification information; And
If the recognition target person is identified in the person recognition processing unit, the control unit further comprises a security response processing process for performing a security response processing operation according to the type and equipment location of the security area to the recognition target recognized through the security response processing unit; Security method using a face recognition-based human feature recognition, characterized in that.
The method of claim 12,
When the control unit determines the location tracking target according to the type of equipment and whether the identification target is identified through the location tracking unit, the control unit determines the location tracking target based on the video received from the video information acquisition unit and the location of the video information acquisition unit where the corresponding video is received. Security tracking method using a face recognition-based human feature recognition further comprises a location tracking process for tracking the location of the location tracking target, and matching or displaying the tracked location to the map data.
The method of claim 12,
If the control unit recognizes the object according to the device type and the device type through the security information notification unit, the control unit detects the detection object detection information, which is information of the illegal recognition object, and the location information of the image information acquisition unit accessed by the illegal recognition object. Security information notification method using the facial feature-based human feature recognition characterized in that it further comprises a security information notification process for notifying the administrator in charge of the security area with reference to the administrator DB containing security information.
The method of claim 12,
The recognition subject identification process,
A person detection step of the control unit detecting and outputting a human image from the image through a person detection unit;
A face detection step of the control unit detecting a face in the human image through a face detection unit and detecting face feature information according to the appearance of the face;
A face recognition step of the control unit identifying a recognition target by comparing the face feature information detected by the face detection unit with the face feature information of the recognition target DB through a face recognition unit;
Extracting human feature information from the human image when the controller fails to identify the person to be recognized through the face recognition unit through the human feature extractor; And
And a feature recognition step of the control unit identifying a person to be recognized by comparing the extracted person appearance feature information and the person appearance feature information pre-registered in the recognition person DB through the feature recognition unit. Security method using feature recognition.
The method of claim 13,
If the video information acquisition unit is a CCTV security terminal,
The controller, when the recognition target person is identified in the security response process, performs a location tracking process in a security response processing operation to track the location of the recognition target. Way.
The method of claim 13,
The image information acquisition unit,
If it's a computer and smart device,
The security response process is
The control unit notifies the computer of authentication success information when the recognition target person is identified through the person recognition processing unit, and allows the computer to be used. If the recognition target person is not identified, the control unit transmits authentication failure information to the computer. Security method using a face recognition-based human feature recognition, characterized in that for performing a security response processing operation to block the use of.

Images