WO2020246652A1 - 얼굴인증을 이용한 게이트 제어 방법 및 시스템 - Google Patents
얼굴인증을 이용한 게이트 제어 방법 및 시스템 Download PDFInfo
- Publication number
- WO2020246652A1 WO2020246652A1 PCT/KR2019/007179 KR2019007179W WO2020246652A1 WO 2020246652 A1 WO2020246652 A1 WO 2020246652A1 KR 2019007179 W KR2019007179 W KR 2019007179W WO 2020246652 A1 WO2020246652 A1 WO 2020246652A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- face
- gate control
- face image
- gate
- Prior art date
Links
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E06—DOORS, WINDOWS, SHUTTERS, OR ROLLER BLINDS IN GENERAL; LADDERS
- E06B—FIXED OR MOVABLE CLOSURES FOR OPENINGS IN BUILDINGS, VEHICLES, FENCES OR LIKE ENCLOSURES IN GENERAL, e.g. DOORS, WINDOWS, BLINDS, GATES
- E06B11/00—Means for allowing passage through fences, barriers or the like, e.g. stiles
- E06B11/02—Gates; Doors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Definitions
- the disclosed technology relates to a method and system for controlling gate opening and closing using face recognition.
- a technology for verifying a person's identity using a face image or performing authentication for a specific procedure is being used.
- a technology for verifying identity by photographing a user's face at an airport with a camera and comparing the feature information of the face image with the feature information extracted from a previously registered passport is used.
- the airport immigration screening is also gradually changed to shorten the time required by the unmanned system automatically processing the immigration screening. According to this trend, it is expected that the immigration procedure will soon be carried out in the direction of controlling the immigration gate based on the minimum information recognition behavior.
- Korean Patent Application Laid-Open No. 10-2016-0017000 name of the invention: face detection apparatus and method
- a technology for detecting a position of a face based on a distance difference image acquired using a stereo camera is disclosed.
- this technology has a problem in that it is possible to determine the location of the object, but cannot determine who the object is specifically.
- the disclosed technology is to provide a method and system for controlling opening and closing of a gate according to a result of face recognition based on an image captured using a plurality of cameras.
- the first aspect of the disclosed technology is the step of acquiring a first face image of a user using an auxiliary terminal connected to a gate control device and transmitting it to a server, a plurality of face images stored in the database by the server Selecting at least one candidate image similar to the first face image from among them, obtaining a second face image for the user using a main terminal connected to the gate control device, and selecting the second face image from among the at least one candidate image.
- the second aspect of the disclosed technology in order to achieve the above technical problem is an auxiliary terminal that captures the first face image to detect a specific user entering the gate among a plurality of users who have entered a photographable area using a plurality of cameras.
- face authentication comprising a server that selects at least one candidate image similar to and a gate control device that matches the at least one candidate image and the second face image received from the server to control a gate according to the matching result It is to provide a gate control system.
- a third aspect of the disclosed technology in order to achieve the above technical problem is an auxiliary terminal that photographs the first face image to detect a specific user entering the gate among a plurality of users who have entered a photographable area using a plurality of cameras.
- a main terminal that photographs second face images for the plurality of users using the same camera as the plurality of cameras, and the first face image among a plurality of face images stored in a database after receiving the first face image
- Providing a gate control system using face authentication comprising a gate control device that selects at least one candidate image similar to and controls the gate according to the matching result by matching the at least one candidate image and the second face image.
- Embodiments of the disclosed technology may have effects including the following advantages. However, since it does not mean that the embodiments of the disclosed technology should include all of them, the scope of the rights of the disclosed technology should not be understood as being limited thereto.
- a gate control method and system using face authentication has an effect of easily processing a user's entry/exit processing by controlling an entry/exit gate using only face authentication.
- FIG. 1 is a flowchart of a gate control method using face authentication according to an embodiment of the disclosed technology.
- FIG. 2 is a block diagram of a gate control system using face authentication according to an embodiment of the disclosed technology.
- FIG. 3 is a diagram illustrating an operation flow of a system according to an embodiment of the disclosed technology.
- FIG. 4 is a diagram showing an auxiliary terminal and a main terminal according to an embodiment of the disclosed technology.
- FIG. 5 is a diagram showing a depth image captured using a 3D depth camera.
- FIG. 6 is a diagram showing a 2D image captured using a 2D camera.
- FIG. 7 is a diagram showing generation of a face image using a depth image and a 2D image.
- Terms such as 1st, 2nd, A, B, etc. may be used to describe various components, but the components are not limited by the above terms, and only for the purpose of distinguishing one component from other components. Is only used.
- a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element.
- the term and/or includes a combination of a plurality of related listed items or any of a plurality of related listed items.
- each of the constituent units to be described below may additionally perform some or all of the functions of other constituent units in addition to its own main function, and some of the main functions of each constituent unit are different. It goes without saying that it may be performed exclusively by. Therefore, the presence or absence of each component described through the present specification will be functionally interpreted.
- FIG. 1 is a flowchart of a gate control method using face authentication according to an embodiment of the disclosed technology.
- the gate control method using face authentication includes the following steps.
- the gate control device acquires a first face image of the user using the auxiliary terminal.
- the gate control device is a device for controlling opening and closing of the gate, and is disposed near the gate to open and close the gate based on the result of the user's face authentication.
- the gate control device is connected to the auxiliary terminal and may acquire a first face image by transmitting a control signal to the auxiliary terminal.
- the auxiliary terminal includes a camera for obtaining a first face image.
- the auxiliary terminal is disposed in front of or on one side of the gate, and may take a picture of a user who enters within a possible distance based on the disposed position.
- the auxiliary terminal uses a 3D camera to capture a depth image of a plurality of users approaching the gate in order to acquire a first face image
- the auxiliary terminal uses a 2D camera to capture a plurality of depth images. Taking a 2D image of the users and detecting the user's position among a plurality of users using distance information obtained from the depth image, and a face corresponding to the detected user's position among the face regions included in the 2D image A step of detecting a region and obtaining the first face image from the face region is performed.
- the auxiliary terminal is equipped with a 3D camera and a 2D camera. Then, to acquire the first face image, the user is photographed using two cameras.
- the auxiliary terminal is installed in front of or on one side of the gate, and a user who enters within a distance that can be photographed based on the installed position can be photographed using two cameras.
- the auxiliary terminal has a 3D camera and a 2D camera at almost the same location, and can take a user by driving both cameras at the same time.
- the auxiliary terminal may capture a depth image of a user using a 3D camera and a 2D image of the same user using a 2D camera.
- the auxiliary terminal may acquire distance information for the user through the depth image and detect the user's face region through the 2D image.
- the auxiliary terminal may acquire a first face image by detecting distance information and a face region of a specific user according to the user's location.
- the auxiliary terminal may be set to determine in advance as a user who wants to detect a user at a location closest to the gate. For example, when three users are photographed in one image, distance information of each user may be obtained, and a specific user having the closest distance information may be detected.
- the location of the user having the closest distance information in the depth image may be identified, and a face region of the user at a location corresponding to the user location identified in the 2D image may be extracted. That is, a first face image may be generated based on extracting a face region for a user at a location closest to the gate.
- the gate control device transmits the generated first face image to the server.
- the first face image generated through the auxiliary terminal is transmitted to the gate control device.
- the gate control device may transmit the first face image to the server.
- the auxiliary terminal may be directly transmitted to the server, but it is necessary for the gate control device to obtain the first face image in advance according to a comparison process with the second face image captured by the main terminal to be described later. Therefore, basically, the gate control device operates to receive the first face image from the auxiliary terminal and transmit it to the server.
- the server selects at least one candidate image similar to the first face image from among the plurality of face images stored in the database.
- the server includes a database storing a plurality of face images.
- the plurality of face images may be images extracted from one of each user's passport, ID card, and boarding pass.
- the server may store a plurality of face information in the database according to the step of storing a plurality of face images prior to step 120.
- the server for example, for One ID-based immigration processing, when multiple passengers register their face information using one of a passport, ID card, and boarding pass in order to process the immigration procedure in advance. Images can be saved in the database in advance.
- the server selects a face image similar to the first face image from among the plurality of face images.
- the server may select at least one face image by extracting feature information representing a unique biometric feature of the user from the first face image and comparing it with feature information of a plurality of face images.
- a threshold value for the similarity may be set in advance to compare whether the similarity between images exceeds the threshold.
- the server may select at least one face image that matches the first face image, and such a face image is called a candidate image.
- the candidate image may or may not be the same user as the user of the first face image. That is, it may be to select face images for candidates estimated to be users of the first face image. This process can primarily filter unnecessary information and reduce the load on the entire system.
- step 130 the gate control device acquires a second face image of the user using the main terminal.
- the gate control device is connected to the main terminal and can acquire a second face image photographed by the main terminal.
- the main terminal acquires a second face image according to the following steps. First, a depth image of a user is photographed using a 3D camera. Then, a 2D image of a user is captured using a 2D camera.
- the main terminal is equipped with a 3D camera and a 2D camera like the auxiliary terminal, and can take a depth image and a 2D image of a user by using them.
- the main terminal detects the user's location and liveness using distance information obtained from the depth image, and detects the user's face area from the 2D image to obtain a second face image.
- the auxiliary terminal and the main terminal photograph the user at different locations.
- the main terminal may be installed in front of the gate. Both terminals can shoot users from their respective installed locations. That is, the gate control device may acquire images of the same user captured at different locations from two terminals.
- the gate control device selects one of the at least one candidate image that matches the second face image. If the previously acquired first face image is used to select a candidate image, the second face image is used to select a specific user from among the candidate images.
- the main terminal may select one candidate image matching the second face image by extracting feature information included in the second face image and comparing it with feature information of at least one candidate image.
- one of the candidate images may be selected by setting a criterion for similarity in advance.
- the gate control device acquires a first face image and a second face image.
- the gate control device compares two face images and detects a similarity between images in order to determine whether the user photographed by the two terminals is the same person. If, as a result of the comparison, it is determined that the users of the two face images are not the same person, the gate control device may transmit a control signal so that the auxiliary terminal, the main terminal, or both terminals may take the face image again. Conversely, if it is determined that they are the same person as a result of comparing the two face images, it can be determined that the photographing for face authentication was performed without errors.
- step 140 the gate control device controls the gate according to the matching result.
- a control signal may be transmitted to open the gate. If the candidate image matching the second face image does not exist, the gate control device may operate to obtain the second face image again by transmitting a control signal to the main terminal.
- a gate control system 200 using face authentication includes an auxiliary terminal 210, a main terminal 220, a server 230, and a gate control device 240.
- the auxiliary terminal 210 captures a first face image for detecting a specific user entering the gate among a plurality of users who have entered a photographable area using a plurality of cameras.
- the auxiliary terminal 210 includes a 3D camera 210a, a 2D camera 210b, an image processor 210c, and a communication module 210d.
- the auxiliary terminal 2100 photographs a depth image of a plurality of users using a 3D camera 210a and a 2D image of the plurality of users using a 2D camera 210b. May control the two cameras 210a and 210b at the same viewpoint to capture a depth image and a 2D image of the user at the same viewpoint.
- the auxiliary terminal 210 obtains distance information for each of a plurality of users from a depth image and uses the image processor 210c to detect a face region in a 2D image.
- a user at the nearest location may be determined as a specific user.
- a face of an area overlapping a location of a specific user among the face areas included in the 2D image may be extracted. That is, it is possible to extract a face of a user in an overlapping area by overlaying a depth image and a 2D image captured using the two cameras 210a and 210b.
- the auxiliary terminal 210 may generate a first image of the user's face according to this operation. The generated first face image is transmitted to the gate control device 240 using the communication module 210d.
- the main terminal 220 photographs a second face image of a plurality of users by using the same plurality of cameras as the auxiliary terminal 210.
- the main terminal 220 includes a 3D camera 220a for photographing depth images of a plurality of users, a 2D camera 220b for photographing 2D images of a plurality of users, and a depth image for each of the plurality of users.
- It includes a processor 220c.
- the main terminal 220 generates a second image of a specific user's face based on a depth image and a 2D image captured from the two cameras 220a and 220b, similarly to the auxiliary terminal generating the first image.
- the second image generated by the main terminal 220 may be transmitted to the gate control device 240 through the communication module 220d.
- the gate control device 240 also includes a communication module 240a like other terminals, and can transmit and receive images using each other's communication modules.
- the server 230 receives the first image generated by the auxiliary terminal 210.
- the server 230 may directly receive the first image from the auxiliary terminal 210 or may receive the first image through the gate control device 240.
- the server 230 stores a plurality of images for comparison with the first image in the database, and may select at least one candidate image similar to the first image from the plurality of images.
- the server extracts feature information included in the first face image and compares the feature information of each of the plurality of face images stored in the database to select at least one face image whose similarity exceeds a threshold value as a candidate image. Can be selected.
- the degree of similarity may be set by an administrator of the system 200.
- the threshold value may be set to 70%, and some face images having a similarity greater than 70% with the feature information of the first image may be selected as candidate images.
- the gate control device 240 matches the at least one candidate image received from the server with the second face image, and controls the gate according to the matching result.
- the gate control device 240 compares the second face image transmitted from the main terminal 220 with at least one candidate image transmitted from the server 230. If there is an image matching the second face image among at least one candidate image, a control signal for opening and closing the gate may be generated. If there is no candidate image matching the second face image, a control signal may be generated to capture a face image of the user again. In this case, a control signal may be transmitted to the auxiliary terminal 210, a control signal may be transmitted to the main terminal 220, or a control signal may be transmitted to each of the two terminals 210 and 220 to retake the user.
- the gate control device 240 may use the feature information to compare the candidate image and the second face image.
- the gate control device 240 may extract feature information included in the second face image and compare the feature information of at least one candidate image to select one candidate image matching the second face image. . According to this process, if there is a candidate image matching the second face image, the gate can be opened. If there is no candidate image matching the second face image, the gate control device 240 may transmit a control signal to at least one of the auxiliary terminal and the main terminal to obtain the face image again.
- the gate control device 240 compares the first face image received from the auxiliary terminal 210 with the second face image received from the main terminal 220 to determine whether the two images are for the same user.
- the gate control device 240 may compare the first face image and the second face image and use the second face image to match at least one candidate image when the similarity between the images exceeds a threshold value. If the first face image and the second face image are compared and the similarity between the images is less than a threshold value, a control signal for rephotographing may be transmitted to at least one of the auxiliary terminal and the main terminal.
- the gate control system 200 may be separated from the server and consist only of an auxiliary terminal, a main terminal, and a gate control device. That is, the gate control device does not transmit the user's first face image acquired from the auxiliary terminal to the server, but can select at least one candidate image by itself.
- the gate control device may include a database synchronized with the database of the server. And when the first face image is received from the auxiliary terminal, some of the plurality of face images stored in the synchronized database may be selected as candidate images. Naturally, a plurality of face images are stored in a synchronized database before the first face image is transmitted.
- FIG. 3 is a diagram illustrating an operation flow of a system according to an embodiment of the disclosed technology.
- the auxiliary terminal 210 and the main terminal 220 may respectively photograph the user's face.
- the gate control device may receive face images from the two terminals 210 and 220, respectively, and may transmit a face image received from the auxiliary terminal 210 among the two face images to the server 230.
- a candidate image for the first face image is received from the server 230, it may be compared with the face image received from the main terminal 220 to determine whether a user exists among the candidate images.
- the server 230 may transmit the candidate image along the communication path 302 previously connected to the gate control device. If the user 301 is a user who has previously registered his or her face image as One ID for immigration processing, there will be a matching image among candidate images. In this case, the gate 240c rotates left and right to open.
- the system 200 may perform authentication using two terminals 210 and 220 provided in the gate 240c.
- a main terminal 220 may be provided in front of the gate 240c, and an auxiliary terminal 210 may be installed on one side of the gate.
- the gate control device 240 may also be disposed at the site where the gate 240c is located along with the two terminals 210 and 220 and maintain a state connected to the two terminals 210 and 220.
- the gate control device 240 may be connected to the two terminals by wire or wirelessly when necessary.
- the two terminals 210 and 220 may be classified into a main terminal and a sub terminal, respectively.
- the operation of the two terminals is the same.
- distance information of a user may be identified through a depth image captured using a 3D camera, and a face region of the user may be extracted through a 2D image captured using a 2D camera.
- two images may be overlaid to detect a user's face in a specific region 401 that overlaps.
- the specific location 401 may be a predetermined point in front of the gate, or a 3D camera and a 2D camera included in the two terminals may be a location in which a user can photograph.
- the user performs face authentication for the purpose of passing through the gate for immigration, and by designating a specific location 401 in advance so that the user's face can be more accurately photographed, the face can be captured more accurately.
- FIG. 5 is a diagram showing a depth image captured using a 3D depth camera.
- a plurality of users photographed in a depth image are displayed in different colors.
- the user 501 on the right is located closer than the user 502 on the left.
- the difference in color between the two user areas means distance information of the user. For example, the brighter the color, the closer it can be determined. Or, the darker the color, the closer it can be seen.
- the size of the user area is not considered and which user is closer to the gate is determined based on distance information.
- FIG. 6 is a diagram showing a 2D image captured using a 2D camera.
- an image captured by a 2D camera includes a user's face. That is, the terminal may extract the face image of each user based on the image captured from the front or the side. Since the 2D image shown in FIG. 6 was photographed at the same time as the depth image of FIG. 5 previously, faces of two users 601 and 602 are photographed. That is, the user face 601 matching the area of the front user 501 and the user face 602 matching the area of the rear user 502 may be photographed together.
- the administrator of the system may determine the optimum point as a specific position to perform face authentication in consideration of the angle of view of the 2D camera, and mark a mark so that the user can accurately stand at that position.
- the contour extraction technique it may be determined that a user is in a position ahead of other users by using the contour extraction technique.
- the accuracy of the distance determination result is significantly reduced. Therefore, the depth image including the distance information and the 2D image photographing the face area of each user are superimposed to obtain the distance information of a specific user. It is desirable to judge.
- FIG. 7 is a diagram showing generation of a face image using a depth image and a 2D image.
- it is possible to determine the entry order of each user's face image by overlaying two images.
- an overlay may be performed by using a depth image having distance information as a base, specifying a region corresponding to a face among the user's images captured in the image, and merging the face image extracted from the front image.
- the overlaid image as described above it is possible to specify which of the face images processed for each user is the face image of a specific user at the first entrance to the gate. For example, after merging face images as described above, it is possible to determine who is the user 701 closest to the gate or who is the user 701 closest to a specific location, and transmit the face image of the user to the server. have. In addition, when face authentication is performed for the user 701, face authentication for the next user 702 may be prepared.
Landscapes
- Engineering & Computer Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Structural Engineering (AREA)
- Civil Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
Abstract
Description
Claims (22)
- 게이트 제어 장치가 연결된 보조단말기를 이용하여 사용자에 대한 제 1 얼굴이미지를 획득하고 서버에 전송하는 단계;상기 서버가 데이터베이스에 저장된 복수개의 얼굴이미지들 중에서 상기 제 1 얼굴이미지와 유사한 적어도 하나의 후보이미지를 선별하는 단계;상기 게이트 제어 장치가 연결된 메인단말기를 이용하여 상기 사용자에 대한 제 2 얼굴이미지를 획득하고 상기 적어도 하나의 후보이미지 중 상기 제 2 얼굴이미지에 매칭되는 하나를 선택하는 단계; 및상기 게이트 제어 장치가 상기 매칭 결과에 따라 게이트를 제어하는 단계;를 포함하는 얼굴인증을 이용한 게이트 제어 방법.
- 제 1 항에 있어서, 상기 서버에 전송하는 단계는,상기 보조단말기가 3D 카메라를 이용하여 상기 게이트에 접근하는 복수의 사용자들에 대한 깊이영상을 촬영하는 단계;상기 보조단말기가 2D 카메라를 이용하여 상기 복수의 사용자들에 대한 2D 영상을 촬영하는 단계; 및상기 깊이영상에서 획득된 거리정보를 이용하여 상기 복수의 사용자들 중 상기 사용자의 위치를 검출하고 상기 2D 영상에 포함된 얼굴영역들 중 상기 검출된 사용자의 위치에 대응되는 얼굴영역을 검출하고 상기 얼굴영역에서 상기 제 1 얼굴이미지를 추출하는 단계;를 포함하는 얼굴인증을 이용한 게이트 제어 방법.
- 제 2 항에 있어서,상기 보조단말기는 상기 거리정보를 토대로 상기 복수의 사용자들 중 상기 게이트에서 가장 가까운 위치의 사용자를 상기 사용자로 검출하는 얼굴인증을 이용한 게이트 제어 방법.
- 제 1 항에 있어서, 상기 후보이미지를 선별하는 단계는,상기 서버가 상기 제 1 얼굴이미지에 포함된 특징정보를 추출하고 상기 데이터베이스에 저장된 상기 복수개의 얼굴이미지들의 특징정보와 비교하여 유사도가 임계값을 초과하는 적어도 하나의 얼굴이미지를 상기 후보이미지로 선별하는 얼굴인증을 이용한 게이트 제어 방법.
- 제 1 항에 있어서, 상기 하나를 선택하는 단계는,상기 메인단말기가 3D 카메라를 이용하여 상기 사용자에 대한 깊이영상을 촬영하는 단계;상기 메인단말기가 2D 카메라를 이용하여 상기 사용자에 대한 2D 영상을 촬영하는 단계; 및상기 깊이영상에서 획득된 거리정보를 이용하여 상기 사용자의 위치 및 라이브니스를 검출하고 상기 2D 영상에서 상기 사용자의 얼굴영역을 검출하여 상기 제 2 얼굴이미지를 추출하는 단계;를 포함하는 얼굴인증을 이용한 게이트 제어 방법.
- 제 1 항에 있어서, 상기 하나를 선택하는 단계는,상기 게이트 제어 장치가 상기 제 2 얼굴이미지에 포함된 특징정보를 추출하고 상기 적어도 하나의 후보이미지들의 특징정보와 비교하여 상기 제 2 얼굴이미지에 매칭되는 하나의 후보이미지를 선택하는 얼굴인증을 이용한 게이트 제어 방법.
- 제 1 항에 있어서, 상기 하나를 선택하는 단계는,상기 게이트 제어 장치가 상기 제 1 얼굴이미지 및 상기 제 2 얼굴이미지를 비교하여 이미지 간의 유사도를 검출하는 단계를 포함하는 얼굴인증을 이용한 게이트 제어 방법.
- 제 1 항에 있어서, 상기 게이트를 제어하는 단계는,상기 제 2 얼굴이미지에 매칭되는 후보이미지가 존재하면 상기 게이트 제어 장치가 제어신호를 전송하여 상기 게이트를 오픈하는 얼굴인증을 이용한 게이트 제어 방법.
- 제 1 항에 있어서, 상기 게이트를 제어하는 단계는,상기 제 2 얼굴이미지에 매칭되는 후보이미지가 존재하지 않으면 상기 게이트 제어 장치가 상기 메인단말기에 제어신호를 전송하여 상기 제 2 얼굴이미지를 다시 획득하는 얼굴인증을 이용한 게이트 제어 방법.
- 제 1 항에 있어서,상기 서버는 상기 제 1 얼굴이미지를 획득하기 이전에 상기 복수개의 얼굴이미지들을 상기 데이터베이스에 저장하고, 상기 복수개의 얼굴이미지들은 복수의 사용자들의 여권, 신분증 및 탑승권 중 적어도 하나로부터 추출된 이미지인 것을 특징으로 하는 게이트 제어 방법.
- 복수개의 카메라를 이용하여 촬영 가능한 영역에 진입한 복수의 사용자들 중 게이트에 입장하는 특정 사용자를 검출하기 위한 첫 번째 얼굴이미지를 촬영하는 보조단말기;상기 복수개의 카메라와 동일한 카메라를 이용하여 상기 복수의 사용자들에 대한 두 번째 얼굴이미지를 촬영하는 메인단말기;상기 첫 번째 얼굴이미지를 수신하고 데이터베이스에 저장된 복수의 얼굴이미지들 중 상기 첫 번째 얼굴이미지와 유사한 적어도 하나의 후보이미지를 선별하는 서버; 및상기 서버에서 수신한 상기 적어도 하나의 후보이미지 및 상기 두 번째 얼굴이미지를 매칭하여 매칭 결과에 따라 게이트를 제어하는 게이트 제어 장치;를 포함하는 얼굴인증을 이용한 게이트 제어 시스템.
- 제 11 항에 있어서, 상기 보조단말기는,상기 복수의 사용자들에 대한 깊이영상을 촬영하는 3D 카메라;상기 복수의 사용자들에 대한 2D 영상을 촬영하는 2D 카메라; 및상기 깊이영상에서 상기 복수의 사용자들 각각에 대한 거리정보를 획득하여 가장 가까운 위치의 사용자를 상기 특정 사용자로 판단하고, 상기 2D 영상에 포함된 얼굴영역들 중 상기 특정 사용자의 위치와 중복되는 영역의 얼굴을 추출하여 상기 첫 번째 이미지를 생성하는 이미지 프로세서;를 포함하는 얼굴인증을 이용한 게이트 제어 시스템.
- 제 11 항에 있어서, 상기 메인단말기는,상기 복수의 사용자들에 대한 깊이영상을 촬영하는 3D 카메라;상기 복수의 사용자들에 대한 2D 영상을 촬영하는 2D 카메라; 및상기 깊이영상에서 상기 복수의 사용자들 각각에 대한 거리정보 및 라이브니스 정보를 획득하여 가장 가까운 위치의 사용자를 상기 특정 사용자로 판단하고, 상기 2D 영상에 포함된 얼굴영역들 중 상기 특정 사용자의 위치와 중복되는 영역의 얼굴을 추출하여 상기 두 번째 이미지를 생성하는 이미지 프로세서;를 포함하는 얼굴인증을 이용한 게이트 제어 시스템.
- 제 11 항에 있어서,상기 서버는 상기 첫 번째 얼굴이미지에 포함된 특징정보를 추출하고 상기 데이터베이스에 저장된 상기 복수의 얼굴이미지들 각각의 특징정보와 비교하여 유사도가 임계값을 초과하는 적어도 하나의 얼굴이미지를 상기 후보이미지로 선별하는 얼굴인증을 이용한 게이트 제어 시스템.
- 제 11 항에 있어서,상기 게이트 제어 장치는 상기 두 번째 얼굴이미지에 포함된 특징정보를 추출하고 상기 적어도 하나의 후보이미지들의 특징정보와 비교하여 상기 제 2 얼굴이미지에 매칭되는 하나의 후보이미지를 선택하는 얼굴인증을 이용한 게이트 제어 시스템.
- 제 11 항에 있어서,상기 게이트 제어 장치는 상기 첫 번째 얼굴이미지 및 상기 두 번째 얼굴이미지를 비교하여 이미지 간의 유사도가 임계값을 초과하면 상기 두 번째 얼굴이미지를 상기 적어도 하나의 후보이미지와 매칭하는 얼굴인증을 이용한 게이트 제어 시스템.
- 제 11 항에 있어서,상기 게이트 제어 장치는 상기 첫 번째 얼굴이미지 및 상기 두 번째 얼굴이미지를 비교하여 이미지 간의 유사도가 임계값 미만이면 상기 보조단말기 및 상기 메인단말기 중 적어도 하나에 재촬영을 위한 제어신호를 전송하는 얼굴인증을 이용한 게이트 제어 시스템.
- 제 11 항에 있어서,상기 게이트 제어 장치는 상기 두 번째 얼굴이미지에 매칭되는 후보이미지가 존재하면 상기 게이트를 오픈하는 얼굴인증을 이용한 게이트 제어 시스템.
- 제 11 항에 있어서,상기 게이트 제어 장치는 상기 두 번째 얼굴이미지에 매칭되는 후보이미지가 존재하지 않으면 상기 보조단말기 및 상기 메인단말기 중 적어도 하나에 제어신호를 전송하여 얼굴이미지를 다시 획득하는 얼굴인증을 이용한 게이트 제어 시스템.
- 제 11 항에 있어서,상기 서버는 상기 첫 번째 얼굴이미지를 수신하기 이전에 상기 상기 데이터베이스에 복수개의 얼굴이미지들을 저장하고, 상기 복수개의 얼굴이미지들은 복수의 사용자들의 여권, 신분증 및 탑승권 중 적어도 하나로부터 추출된 이미지인 것을 특징으로 하는 게이트 제어 시스템.
- 복수개의 카메라를 이용하여 촬영 가능한 영역에 진입한 복수의 사용자들 중 게이트에 입장하는 특정 사용자를 검출하기 위한 첫 번째 얼굴이미지를 촬영하는 보조단말기;상기 복수개의 카메라와 동일한 카메라를 이용하여 상기 복수의 사용자들에 대한 두 번째 얼굴이미지를 촬영하는 메인단말기; 및상기 첫 번째 얼굴이미지를 수신하고 데이터베이스에 저장된 복수의 얼굴이미지들 중 상기 첫 번째 얼굴이미지와 유사한 적어도 하나의 후보이미지를 선별하고, 상기 적어도 하나의 후보이미지 및 상기 두 번째 얼굴이미지를 매칭하여 매칭 결과에 따라 게이트를 제어하는 게이트 제어 장치;를 포함하는 얼굴인증을 이용한 게이트 제어 시스템.
- 제 21 항에 있어서,상기 게이트 제어 장치의 데이터베이스는 서버의 데이터베이스와 동기화되고, 상기 첫 번째 얼굴이미지를 수신하기 이전에 상기 서버의 데이터베이스에 저장된 복수개의 얼굴이미지들을 수신하는 게이트 제어 시스템.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020190067147A KR102080252B1 (ko) | 2019-06-07 | 2019-06-07 | 얼굴인증을 이용한 게이트 제어 방법 및 시스템 |
KR10-2019-0067147 | 2019-06-07 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020246652A1 true WO2020246652A1 (ko) | 2020-12-10 |
Family
ID=69670975
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2019/007179 WO2020246652A1 (ko) | 2019-06-07 | 2019-06-14 | 얼굴인증을 이용한 게이트 제어 방법 및 시스템 |
Country Status (2)
Country | Link |
---|---|
KR (1) | KR102080252B1 (ko) |
WO (1) | WO2020246652A1 (ko) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102487033B1 (ko) * | 2020-10-27 | 2023-01-12 | 주식회사 미르메이트 | 인공 지능 기반 영상 처리 장치 |
KR102555103B1 (ko) * | 2021-09-02 | 2023-07-17 | (주)씨유박스 | 얼굴영상을 이용한 액티브 라이브니스 검출 방법 및 장치 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20030091395A (ko) * | 2002-05-27 | 2003-12-03 | 주식회사 케이디엔스마텍 | 생체인식여권 및 사증을 이용한 출입국 자동관리시스템 |
KR20110006971A (ko) * | 2009-07-15 | 2011-01-21 | 주식회사 아이디테크 | 출입통제시스템의 출입인증 사용자에 대한 출입허가용 정면얼굴 영상 검출 방법 |
KR101100700B1 (ko) * | 2011-06-03 | 2011-12-29 | 주식회사 테스나 | 생체인식여권을 이용한 출입국 관리시스템 |
KR20140031105A (ko) * | 2012-09-03 | 2014-03-12 | 삼성전자주식회사 | 인식 대상으로부터 3차원 거리 정보 추출하는 방법 및 이를 위한 장치 |
KR20170077366A (ko) * | 2015-12-28 | 2017-07-06 | 삼성에스디에스 주식회사 | 얼굴 인식 시스템 및 방법 |
-
2019
- 2019-06-07 KR KR1020190067147A patent/KR102080252B1/ko active IP Right Grant
- 2019-06-14 WO PCT/KR2019/007179 patent/WO2020246652A1/ko active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20030091395A (ko) * | 2002-05-27 | 2003-12-03 | 주식회사 케이디엔스마텍 | 생체인식여권 및 사증을 이용한 출입국 자동관리시스템 |
KR20110006971A (ko) * | 2009-07-15 | 2011-01-21 | 주식회사 아이디테크 | 출입통제시스템의 출입인증 사용자에 대한 출입허가용 정면얼굴 영상 검출 방법 |
KR101100700B1 (ko) * | 2011-06-03 | 2011-12-29 | 주식회사 테스나 | 생체인식여권을 이용한 출입국 관리시스템 |
KR20140031105A (ko) * | 2012-09-03 | 2014-03-12 | 삼성전자주식회사 | 인식 대상으로부터 3차원 거리 정보 추출하는 방법 및 이를 위한 장치 |
KR20170077366A (ko) * | 2015-12-28 | 2017-07-06 | 삼성에스디에스 주식회사 | 얼굴 인식 시스템 및 방법 |
Also Published As
Publication number | Publication date |
---|---|
KR102080252B1 (ko) | 2020-02-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR100682898B1 (ko) | 적외선을 이용한 영상 장치 및 그의 영상 식별 방법 | |
US8422746B2 (en) | Face authentication system and authentication method thereof | |
WO2017069385A1 (ko) | 모바일 디바이스를 이용한 신분증 인식 장치 | |
WO2020246652A1 (ko) | 얼굴인증을 이용한 게이트 제어 방법 및 시스템 | |
WO2017099427A1 (ko) | 손가락 마디 및 지정맥 기반의 융합형 생체 인증 방법 및 그 장치 | |
WO2018012928A1 (ko) | 얼굴 인식을 이용한 사용자 인증 방법 및 그 장치 | |
KR101096984B1 (ko) | 촬영 장치, 촬영 방법 및 촬영 프로그램을 기록한 컴퓨터 판독가능한 기록 매체 | |
CN104809794B (zh) | 门禁控制方法及系统 | |
US7190814B2 (en) | Image comparison apparatus and method for checking an image of an object against a stored registration image | |
WO2017115905A1 (ko) | 인체 포즈 인지 시스템 및 방법 | |
JP5001195B2 (ja) | 入室管理システム | |
CN108764053A (zh) | 图像处理方法、装置、计算机可读存储介质和电子设备 | |
WO2021172833A1 (ko) | 물체 인식 장치, 물체 인식 방법 및 이를 수행하기 위한 컴퓨터 판독 가능한 기록 매체 | |
WO2021145502A1 (ko) | 얼굴 및 스트레스 인식 장치 및 방법 | |
US20220245978A1 (en) | Gate system, gate apparatus, control method therefor, and program | |
WO2022045507A1 (ko) | 게이트 출입을 위한 인증 방법 및 장치 | |
KR101596363B1 (ko) | 얼굴 인식에 의한 출입관리장치 및 방법 | |
JP2007249298A (ja) | 顔認証装置および顔認証方法 | |
JP5297397B2 (ja) | ドアセキュリティシステムおよびその制御装置 | |
WO2019027103A1 (ko) | 시선 인식을 이용한 잠금해제 장치 및 방법 | |
WO2020246651A1 (ko) | 복수의 카메라를 이용한 얼굴인증 방법, 단말기 및 시스템 | |
JP5001194B2 (ja) | 入室管理システム | |
WO2022019601A1 (ko) | 영상의 객체 특징점 추출과 이를 이용한 영상검색 시스템 및 방법 | |
KR102518061B1 (ko) | 얼굴 윤곽 추정을 통한 마스크 착용 여부 확인 방법 및 장치 | |
WO2021010557A1 (ko) | 위험 감지 시스템 및 방법 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19931569 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19931569 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 26.04.2022) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19931569 Country of ref document: EP Kind code of ref document: A1 |