KR200240824Y1 - security apparatus using data of face photographing - Google Patents

Abstract

The present invention relates to a security device using data photographing a person's face. According to the present invention, a video tracking camera is installed in a public place to photograph people's faces. The face data photographed in this way is compared with the key character data of criminals, etc. previously stored in the database of the central control unit, so that the key character such as criminals can be detected more quickly and reliably.

Description

Security apparatus using data of face photographing}

The present invention relates to a security device, and more particularly to a security device through the face recognition of a person.

Currently, due to the development of security and authentication methods and devices, security devices using fingerprint recognition, iris recognition, vein recognition, voice recognition, etc. are generally installed in places that control access of the general public or places requiring security authentication. do. However, in the security device described above, the authentication target person has to recognize his or her fingerprint, iris, vein, voice, etc. in the recognizer, which is cumbersome. In addition, in case of intentional intrusion, this recognition process can be intentionally avoided, which makes it difficult to control the external intrusion.

In addition, in places where surveillance of impure persons such as financial institutions and government offices are to be carried out, CCTVs are usually installed to monitor the impure persons or to record the images taken by the CCTV on a videotape. When an incident occurred, I would play the tape of the time zone where the accident or incident occurred and find the main mother of the accident or incident. Therefore, there was no way to find out in advance the person of interest or wanted criminal among those who visit the place with unintended intention and purpose.

Accordingly, an object of the present invention is to provide a security device that can solve the above problems.

Another object of the present invention is to provide a security device that does not require an intentional authentication process.

Another object of the present invention is to analyze a person's photograph of a person of interest or a wanted person in advance and store it in the database of the security device and compare the picture image taken by the surveillance camera with the data stored in advance. It provides a method and a device that can find in real time immediately.

In order to achieve the above object, according to the present invention, a non-specified number of faces, the camera unit comprising a camera, an image caption unit, a camera driver and a camera drive control unit; A management computer unit to which face data photographed by the camera unit is transmitted; A preprocessor for preprocessing the image data; A feature point detector for detecting feature points in the preprocessed video signal; A three-dimensional information extraction unit for extracting three-dimensional information of the image; A mask generator for generating a mask based on the extracted three-dimensional information; A shape recognition unit recognizing the generated mask; It is composed of a database unit for storing the image data of the person or wanted criminals, the central control to connect the management computer unit and the network receives the face data of the camera unit from the management computer unit to extract the main characters needing control It provides a security device characterized in that it comprises a wealth.

1 is a schematic configuration diagram for explaining a security device according to an embodiment of the present invention.

<Explanation of symbols for the main parts of the drawings>

10a to 10d: Video tracking camera section 11: Video caption section

12: Camera 13: Camera drive part

14: camera drive control unit 20: management computer

30: central control part 31: pretreatment part

32: feature detection unit 33: three-dimensional information extraction unit

34: database portion 35: shape recognition portion

36: mask generation unit

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention will be described in detail.

1 is a schematic configuration diagram for explaining a security device according to an embodiment of the present invention.

Referring to the drawings, airport tracking rooms, public places where a lot of people gather, and the image tracking camera units 1,2,3,4 (10a, 10b, 10c, 10d) are installed in the public spaces, plazas, and terminals. The image tracking camera is a web camera capable of transmitting data on a network, and may be installed in accordance with a situation such as an area of an installation site, and in some cases, may be mounted on a rail. The face data photographed by the image tracking camera is transmitted to the management computer 20, which is in turn transmitted to the central control unit 30 through a network. Although not shown, the image tracking camera unit controls an image caption unit 11 for capturing an image captured by the camera 12, a camera driver 13 driving the camera 12, and the camera driver 13. And a camera drive control unit 14. In addition, the central control unit 30 is a pre-processing unit 31 for processing an image captioned by the camera 12, a feature point detector 32 for detecting a feature point in the image and extracting three-dimensional information from the image A mask generator 36 for generating a mask based on the information extracted by the 3D information extractor 33 and the 3D information extractor 33, and a shape recognizer 35 that recognizes a shape based on the mask. And a memory and a database section 34.

In this case, the face picture transmitted to the central control unit 30 is transmitted in units of frames, and the face picture transmitted in units of frames is recognized as an object in the following manner.

First, the image capturing unit 11 captures an image captured by the cameras 1-4 of the image tracking camera units 10a-10d as image input units by the image capturing unit 11 so as to form a single screen. 30, and the transmitted image signal is subjected to limbal line correction and filtering by the preprocessor 31, and the image signal processed by the preprocessor 31 is input to the feature point detector 32 to Comparing the full screen with the current screen, a candidate region to be examined is generated from an image signal input by using motion data, a feature point is detected from the generated candidate region, and the detected feature point is used by the camera. Rearrange the input image.

The detected feature points and rearranged image data are input to the 3D information extracting unit 33 and the camera driving control unit 14, respectively, and are immediately inputted from the camera driving control unit 14 to the process described above. Compute the moving distance of the candidate area by comparing with the inspection candidate area data processed through the camera, determine the rotation direction and the rotation angle of the camera so that the object can be re-photographed by rotating the camera by the moved distance, and the camera driver (13) is operated to rotate the camera 12 in the predetermined direction by the angle of rotation described above.

The feature point detector 32 receives the generated candidate region data, the feature points, and the rearranged image data, rearranges the image using the set feature points, and then sets the feature points of the rearranged image in advance. Determining whether the face is a human face by comparing the feature points, and if the face is a human face, rearrange the images, generate three-dimensional information by using displacement information between multiple images, and generate masks from the generated three-dimensional information. A new mask that is different for each individual is generated in comparison with the reference mask stored in the section 36.

The new mask is compared with the mask data stored in the database 34 to determine whether the new mask exists in the database 34. If the same mask data exists in the database 34 as a result of the determination, the new mask is recognized as a registered mask. If the registered mask is a wanted criminal or a person of interest, an alarm is issued. If it is determined that it does not exist in the database, it is recognized as an unregistered mask and does not emit an alarm.

In this way, the central control unit 30 compares and examines the faces of the subjects previously stored in the database 34 and the faces photographed and transmitted through the image tracking camera. When the data stored in the database 34 and the data transmitted through the image tracking camera coincide with each other, the person who is photographed by the image tracking camera is arrested by adding a suppressor or mobilizing the suppression equipment.

As such, in the present invention, by tracking an important person such as a wanted person input to the database 34 of the central control unit 30 through the image tracking camera and taking an emergency contact, the criminal can be arrested more quickly and effectively. have. In addition, since the image tracking camera can recognize a face even when disguised or disguised, it is possible to more reliably extract and arrest an important subject such as a wanted person.

On the other hand, the above-described security device can be applied to places where only authorized persons can enter and exit, for example, research institutes conducted in secret and military operations in military operations. That is, after storing the face data of the authenticated people who have been previously allowed access to the database of the central control unit, the face data photographed through the image tracking camera are compared with each other. As a result of the comparison, if the person who does not have the face data input to the database of the central control unit is not allowed to enter, the security can be more reliably maintained.

In addition, the face (mask) of the person of interest or the wanted criminals inputted into the database of the computer system of the central control unit is input in advance in the shape according to various situation settings. In other words, in case a person in need of disguise or disguise a wig or a beard, set the situation in which the face or disguise of a person in need of disguise or disguise it as a camouflage or disguise such as a wig or a beard. If a person or wanted criminal is disguised or disguised with a wig, beard, or hat, it can be easily written. In addition, the security device is equipped with an artificial intelligence system by the self-learning of the system itself, and has the ability to catch people of interest or wanted criminals even if a situation other than the situation entered in advance occurs.

Although the present invention described above is described according to the drawings and limited, for example, various changes and modifications are possible to those skilled in the art without departing from the technical spirit of the present invention according to the matter. to be.

As described above, according to the present invention, the key figures such as criminals can be quickly and reliably compared by comparing the person data photographed using the image tracking camera with the key figures data such as criminals stored in the database of the central control unit. You can be arrested.

Claims (4)

A camera unit configured to photograph an unspecified number of faces and include a camera, an image caption unit, a camera driver, and a camera drive controller; A management computer unit to which face data photographed by the camera unit is transmitted; A preprocessor for preprocessing the image data; A feature point detector for detecting feature points in the preprocessed video signal; A three-dimensional information extraction unit for extracting three-dimensional information of the image; A mask generator for generating a mask based on the extracted three-dimensional information; A shape recognition unit recognizing the generated mask; It is composed of a database unit for storing the image data of the person or wanted criminals, the central control to connect the management computer unit and the network receives the face data of the camera unit from the management computer unit to extract the main characters needing control Security device characterized in that it comprises a wealth. The security device of claim 1, wherein the camera unit is a web camera capable of transmitting data on a network. The security device of claim 1, wherein the camera is installed to be movable by attaching to a rail. According to claim 1, The face data photographed through the camera unit is compared with each other and the face data of the key subjects that need to be controlled in advance in the database of the central control unit or the face data of the authorized member authorized to access Guard device characterized in that.