KR950029985A - Personal recognition method using image processing of human face - Google Patents

Personal recognition method using image processing of human face Download PDF

Info

Publication number
KR950029985A
KR950029985A KR1019940007903A KR19940007903A KR950029985A KR 950029985 A KR950029985 A KR 950029985A KR 1019940007903 A KR1019940007903 A KR 1019940007903A KR 19940007903 A KR19940007903 A KR 19940007903A KR 950029985 A KR950029985 A KR 950029985A
Authority
KR
South Korea
Prior art keywords
image
pixels
window
predetermined
value
Prior art date
Application number
KR1019940007903A
Other languages
Korean (ko)
Other versions
KR960013819B1 (en
Inventor
송병탁
신지 오자와
Original Assignee
배순훈
대우전자 주식회사
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 배순훈, 대우전자 주식회사 filed Critical 배순훈
Priority to KR1019940007903A priority Critical patent/KR960013819B1/en
Publication of KR950029985A publication Critical patent/KR950029985A/en
Application granted granted Critical
Publication of KR960013819B1 publication Critical patent/KR960013819B1/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172Classification, e.g. identification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/174Facial expression recognition
    • G06V40/175Static expression
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/18Eye characteristics, e.g. of the iris
    • G06V40/19Sensors therefor

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Ophthalmology & Optometry (AREA)
  • Collating Specific Patterns (AREA)

Abstract

본 발명은 연속으로 입력되는 얼굴의 화상으로부터 눈깜빡임을 검출하여 일정크기의 영역인 기준위치를 결정하고 이로부터 모자이크 패턴을 생성하여 컴퓨터 내부에 미리 저장, 등록된 사람의 모자이크 패턴과 비교함으로써 동일인 여부를 식별하는 컴퓨터의 화상처리를 이용한 개인 인식 방법에 관한 것이다.The present invention detects eye blinks from images of faces that are continuously input, determines a reference position, which is an area of a predetermined size, generates a mosaic pattern therefrom, and compares them with a mosaic pattern of a person who is previously stored and registered in a computer. The present invention relates to a personal recognition method using image processing of a computer for identifying an ID.

등록된 사람의 일정특징을 저장하여 컴퓨터를 이용한 화상처리로서 동일인여부를 식별하는 인식방법에 있어서, 본 발명에 따른 인식방법은 연속으로 입력되는 얼굴의 화상으로부터 일정크기의 영역인 기준위치를 구하여 모자이크 패턴을 생성시키는 중요특징 추출단계(50), 및 상기 중요 특징 추출단계에서 생성된 모자이크 패턴과 모자이크 파일에 저장된 모자이크 패턴을 비교하여 개인을 인식하는 인식단계(100)로 구성되며, 상기 일정 특징은 사람 얼굴의 일정영역을 추출하여 생성된 모자이크 패턴인 것을 특징으로 한다.In a recognition method for storing a certain feature of a registered person and identifying whether the same person is an image processing using a computer, the recognition method according to the present invention obtains a reference position, which is an area of a constant size, from an image of a face that is continuously inputted, and performs a mosaic. An important feature extraction step 50 for generating a pattern, and a recognition step 100 for recognizing an individual by comparing the mosaic pattern generated in the important feature extraction step and the mosaic pattern stored in the mosaic file, the predetermined feature is Characterized in that it is a mosaic pattern generated by extracting a predetermined region of the human face.

Description

사람 얼굴의 화상처리를 이용한 개인 인식방법Personal recognition method using image processing of human face

본 내용은 요부공개 건이므로 전문내용을 수록하지 않았음Since this is an open matter, no full text was included.

제1도는 본 발명에 따른 사람 얼굴의 화상처리를 이용한 개인 인식 방법의 전단계를 개략적으로 보인 흐름도, 제2도는 본 발명에 따른 사람 얼굴의 화상처리를 이용한 개인 인식 방법에 있어서 기준위치를 추출하는 단계를 보인 흐름도, 제3도는 본 발명에 따른 사람 얼굴의 화상처리를 이용한 개인 인식 방법에 있어서 눈의 영역인 일정영역을 추출하는 단계를 보인 흐름도, 제4도는 본 발명에 따른 사람 얼굴의 화상처리를 이용한 개인 인식 방법에 있어서 눈의 좌표를 추출하는 단계를 보인 흐름도.1 is a flow chart schematically showing the previous step of the personal recognition method using the image processing of the human face according to the present invention, Figure 2 is a step of extracting a reference position in the personal recognition method using the image processing of the human face according to the present invention Figure 3 is a flow chart showing the step of extracting a predetermined area which is the eye area in the personal recognition method using the image processing of the human face according to the present invention, Figure 4 is a flow chart showing the image processing of the human face according to the present invention Flow chart showing a step of extracting the coordinates of the eye in the personal recognition method.

Claims (8)

등록된 사람의 일정특징을 저장하여 컴퓨터를 이용한 화성처리로서 동일인 여부를 식별하는 인식방법에 있어서, 연속으로 입력되는 얼굴의 화상으로부터 일정크기의 영역인 기준위치를 구하여 모자이크 패턴을 생성시키는 중요특징 추출단계(50), 및 상기 중요 특징 추출단계에서 생성된 모자이크 패턴과 저장된 모자이크 패턴을 비교하여 개인을 인식하는 인식단계(100)로 구성되며, 상기 일정특징은 사람 열굴의 일정영역을 추출하여 생성된 모자이크 패턴인 것을 특징으로 하는 사람 얼굴의 화상 처리를 이용한 개인 인식 방법.In a recognition method of storing a predetermined feature of a registered person and identifying whether it is the same as a computer processing process, extracting an important feature by generating a mosaic pattern by obtaining a reference position, which is an area of a constant size, from an image of a face that is continuously input. And a recognition step 100 for recognizing the individual by comparing the mosaic pattern generated in the important feature extraction step with the stored mosaic pattern, wherein the predetermined feature is generated by extracting a predetermined region of the human crater. A personal recognition method using image processing of a human face, which is a mosaic pattern. 제1항에 있어서, 상기 중요 특징 추출단계(50)의 일정크기의 영역인 기준위치는, 연속으로 입력되는 얼굴 화상에 소버 연산자(Sober operator)를 사용하여 에지화상을 추출하는 단계(12), 연속으로 입력되는 상기 에지 화상의 두 프레임을 취하여 감산하여 상기 두프레임중 한 프레임에 눈깜빡거림이 있을때 눈깜빡거림 영역의 화상에서 0 보다 큰 값을 차분화상을 얻는 프레임 감산단계(20), 차분화상을 2진화상으로 변환하는 단계(25), 눈깜빡거림 영역의 화소만이 선택되도록 불필요한 화소를 제거하는 7×7 화소필터로 상기 2진 화상을 필터링하여 필터링된 2진 화상을 얻는 필터링 단계(26, 28), 상기 필터링된 2진 화상을 고정된 크기의 윈도우로 스캔하여 눈의 영역으로서 일정영역을 추출하는 단계(29), 상기 일정영역을 수평 및 수직방향으로 사영(projection)하며 그 사영을 추적하여 눈의 좌표를 추출하는 단계(31), 상기 눈의 좌표를 사용하여 일정크기의 영역을 추출하는 단계(32)로 결정되는 것을 특징으로 하는 사람 얼굴의 화상처리를 이용한 개인 인식 방법.The method according to claim 1, wherein the reference position, which is an area of a predetermined size of the important feature extraction step 50, extracts an edge image by using a Sober operator on a face image that is continuously input (12), Frame subtraction step 20, which takes two frames of the edge image continuously inputted and subtracts them to obtain a differential image with a value greater than 0 in the image of the blinking region when one of the two frames has blinking. Converting an image to a binary image (25); filtering the binary image to obtain a filtered binary image by filtering the binary image with a 7x7 pixel filter that removes unnecessary pixels so that only pixels in the blinking region are selected (26, 28), scanning the filtered binary image with a fixed size window to extract a predetermined area as an area of the eye (29), projecting the predetermined area in a horizontal and vertical direction; Personal recognition using image processing of the human face, characterized in that the step of extracting the coordinates of the eye by tracking the projection (31), the step of extracting a predetermined size region using the coordinates of the eye (32). Way. 제2항에 있어서, 일정크기의 영역인 기준위치는 75×75 화소로 구성되며, 머리카락 부분을 포함하지 않는 것을 특징으로 하는 사람 얼굴의 화상처리를 이용한 개인 인식 방법.The personal recognition method using image processing of a human face according to claim 2, wherein the reference position, which is an area of a predetermined size, is composed of 75 x 75 pixels and does not include a hair part. 제2항에 있어서, 상기 프레임은 30프레임/초의 속도로 입력되는 프레임 가운데 10개의 프레임 간격으로 얻어지는 것을 특징으로 하는 사람 얼굴의 화상처리를 이용한 개인 인식 방법.The method of claim 2, wherein the frame is obtained at an interval of 10 frames among frames input at a rate of 30 frames / second. 제2항에 있어서, 눈의 영역으로서 일정영역을 추출하는 단계(29)는, a) 고정된 크기의 윈도우를 설정하여 상기 2진 화상에 적용시키는 단계(41), b) 윈도우내의 화소를 더하여 윈도우 값을 구하는 단계(42), c) 상기 화소를 더한 윈도우 값을 제1소정값과 비교하는 단계(43), d) 상기 윈도우 값이 제1소정값보다 작을때 잡음으로 판단하고 화상의 끝인가를 판단하며 화상의 끝이 아닐때 b)단계를 반복시키는 단계(44), e) 화상의 끝일때 다음 프레임의 필터링된 2진 화상을 얻는 필터링 단계(26)부터 반복시키는 단계(45), f) 상기 윈도우 값이 제1소정값보다 클때 오른쪽 눈 영역으로 판단하고 왼쪽눈 영역을 추출하기 위하여 윈도우의 크기를 4배로 하는 단계(46), g) 상기 f)단계의 윈도우 내의 모든 화소를 더한 윈도우 값을 제2소정값과 비교하는 단계(47), h)상기 윈도우 값이 제2소정값보다 클때 왼쪽눈 영역으로 판단하고 양쪽 눈에 두배의 윈도우를 적용하는 단계(48), i)상기 윈도우 값이 0이거나 제2소정값보다 작을때 상기 f)단계 판단은 잘못된 것으로 하고 화상의 끝인지를 판단하는 단계(49), j)화상의 끝일때 다음 프레임의 필터링된 2진 화상을 얻는 필터링 단계(26)부터 반복시키며 화상의 끝이 아닐때 상기(g)단계부터 반복시키는 단계(60)로 구성되는 것을 특징으로 하는 사람 얼굴의 화상처리를 이용한 개인 인식 방법.The method of claim 2, wherein the step 29 of extracting a predetermined area as an area of the eye comprises: a) setting a window of fixed size and applying it to the binary image (41), b) adding pixels in the window; Obtaining a window value (42); c) comparing the window value obtained by adding the pixel with a first predetermined value (43); and d) determining that the window value is less than the first predetermined value and determining that it is noise and ending the image. (44) repeating step b) when it is not the end of the picture, e) filtering step 26 to obtain a filtered binary picture of the next frame at the end of the picture, f) determining the right eye area when the window value is greater than the first predetermined value and quadrupling the size of the window to extract the left eye area (46), g) adding all the pixels in the window of step f) Comparing a window value with a second predetermined value (47), h) said window (48) Applying a double window to both eyes when the second predetermined value is greater than the second predetermined value, i) When the window value is 0 or smaller than the second predetermined value, the determination of step f) is incorrect. And determining whether the end of the image is the end (49), j) repeating from the filtering step 26 of obtaining a filtered binary image of the next frame at the end of the image, and repeating from the step (g) if not the end of the image. Personal recognition method using the image processing of the human face, characterized in that it comprises a step (60). 제2항에 있어서, 눈의 좌표를 추출하는 단계(31)는 a)상기 두배의 윈도우가 적용된 일정영역의 화상을 수평 사영(projection)하는 단계(51), b)밀접하게 위치된 화소의 세트수가 2개 인가를 판단하여 양쪽눈이 추출되었는가를 판단하는 양쪽눈 추출 단계(52), c)화소의 세트수가 1개 일때 양쪽눈이 추출되지 않은 것으로 판단하여 다음 프레임의 필터링된 2진 화상을 얻는 필터링 단계(26)부터 반복시키는 단계(53), d)화소의 세트수가 2개일때 양쪽눈이 추출된 것으로하여 화소의 세트중심을 계산함으로써 수평축의 좌표(X1, X2)를 정하는 단계(54), e)왼쪽 및 오른쪽 눈에 대하여 수직 사영하는 단계(55), f)화소의 세트중심을 계산함으로써 수직축의 좌표(Y1, Y2)를 계산하는 단계(56)로 구성되는 것을 특징으로 하는 사람 얼굴의 화상처리를 이용한 개인 인식 방법.The method of claim 2, wherein the step of extracting the coordinates of the eyes (31) comprises the steps of: a) horizontally projecting an image of a certain region to which the double window is applied (b), b) a set of closely located pixels The binocular extraction step 52 for judging whether two eyes are extracted by judging whether the number is two, c) when the set number of pixels is one, it is determined that both eyes are not extracted and the filtered binary image of the next frame is obtained. (53) repeating from the filtering step 26 to be obtained; and d) determining coordinates (X1, X2) on the horizontal axis by calculating the set center of the pixels with both eyes being extracted when the number of pixels is two (54). ), e) a step of projecting vertically with respect to the left and right eyes (55), and f) calculating (56) the coordinates (Y1, Y2) of the vertical axis by calculating the set center of the pixels. Personal recognition method using face image processing. 제1항 또는 제3항에 있어서, 상기 모자이크 패턴은, 상기 추출된 눈의 좌표를 사용하여 75×75 화소의 고정된 크기인 기준위치로 얼굴영역을 특징짓는 단계(57), 상기 75×75 화소의 고정된 크기를 3×3 화소의 블럭으로 나누고 그 3×3 화소내의 화소값을 평균함으로써 하나의 값을 갖도록 하여 25×25 불럭으로 구성하여 생성하는 단계(58)로 추출되는 것을 특징으로 하는 사람 얼굴의 화상처리를 이용한 개인 인식 방법.The method of claim 1 or 3, wherein the mosaic pattern is characterized by using the extracted eye coordinates to define a facial region at a reference position having a fixed size of 75 × 75 pixels (57). Dividing the fixed size of the pixel into blocks of 3x3 pixels and averaging the pixel values in the 3x3 pixels so as to have a single value so as to produce 25x25 blocks, which is extracted (58). Personal recognition method using image processing of person's face. 제1항에 있어서, 상기 개인 인식 단계(100)는 a)중요 특징 추출단계로 부터 생성된 모자이크 패턴과 저장된 모자이크 패턴 간에 유우클리드 거리를 계산하여 최소값을 선택하는 단계(80), b)상기 유우클리드 거리 계산의 최소값과 제3소정값을 비교하는 단계(81), c)상기 유우클리드 거리 계산의 최소값이 제3소정값보다 크면 동일인이 아닌 것으로 판단하여 등록되지 않은 사람으로 표시하는 단계(82), d) 상기 유우클리드 거리계산의 최소값이 제3소정값보다 작으면 동일인으로 판단하여 이름을 표시하는 단계(83)로 구성된 것을 특징으로 하는 사람 얼굴의 화상처리를 이용한 개인 인식 방법.The method of claim 1, wherein the personal recognition step (100) comprises the steps of: a) selecting the minimum value by calculating the Euclidean distance between the mosaic pattern generated from the important feature extraction step and the stored mosaic pattern (80), and b) the dairy cow Comparing the minimum value of the cleat distance calculation and the third predetermined value (81), c) if the minimum value of the Euclidean distance calculation is greater than the third predetermined value, it is determined that the person is not the same person and displayed as an unregistered person (82) and d) if the minimum value of the Euclidean distance calculation is smaller than a third predetermined value, determining that the same person is the same and displaying a name (83). ※ 참고사항 : 최초출원 내용에 의하여 공개하는 것임.※ Note: The disclosure is based on the initial application.
KR1019940007903A 1994-04-13 1994-04-13 Personal identification by image processing human face of series of image KR960013819B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1019940007903A KR960013819B1 (en) 1994-04-13 1994-04-13 Personal identification by image processing human face of series of image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1019940007903A KR960013819B1 (en) 1994-04-13 1994-04-13 Personal identification by image processing human face of series of image

Publications (2)

Publication Number Publication Date
KR950029985A true KR950029985A (en) 1995-11-24
KR960013819B1 KR960013819B1 (en) 1996-10-10

Family

ID=19381093

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1019940007903A KR960013819B1 (en) 1994-04-13 1994-04-13 Personal identification by image processing human face of series of image

Country Status (1)

Country Link
KR (1) KR960013819B1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20020038162A (en) * 2000-11-16 2002-05-23 구자홍 Iris recognition method for iris recognition system using both of the eyes
KR100445800B1 (en) * 2001-01-05 2004-08-25 학교법인 인하학원 Face-image recognition method of similarity measure using correlation
WO2006080755A1 (en) * 2004-10-12 2006-08-03 Samsung Electronics Co., Ltd. Method, medium, and apparatus for person-based photo clustering in digital photo album, and person-based digital photo albuming method, medium, and apparatus

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20180065527A (en) * 2016-12-08 2018-06-18 경창산업주식회사 Vehicle side-rear warning device and method using the same

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20020038162A (en) * 2000-11-16 2002-05-23 구자홍 Iris recognition method for iris recognition system using both of the eyes
KR100445800B1 (en) * 2001-01-05 2004-08-25 학교법인 인하학원 Face-image recognition method of similarity measure using correlation
WO2006080755A1 (en) * 2004-10-12 2006-08-03 Samsung Electronics Co., Ltd. Method, medium, and apparatus for person-based photo clustering in digital photo album, and person-based digital photo albuming method, medium, and apparatus

Also Published As

Publication number Publication date
KR960013819B1 (en) 1996-10-10

Similar Documents

Publication Publication Date Title
JP4307496B2 (en) Facial part detection device and program
Morris et al. Blink detection for real-time eye tracking
EP0984386B1 (en) Method of detecting a human face
CA2236082A1 (en) Method and apparatus for detecting eye location in an image
KR20010103631A (en) System and method for biometrics-based facial feature extraction
US20060281969A1 (en) System and method for operation without touch by operators
JP4739870B2 (en) Sunglasses detection device and face center position detection device
CN105893963B (en) A kind of method of the best frame easy to identify of single pedestrian target in screening video
KR20040059313A (en) Method of extracting teeth area from teeth image and personal identification method and apparatus using teeth image
JP2008113902A (en) Apparatus and method for detecting eye opening degree
CN115039150A (en) Determination method, determination device, and determination program
CN111047619B (en) Face image processing method and device and readable storage medium
KR950029985A (en) Personal recognition method using image processing of human face
KR20100121817A (en) Method for tracking region of eye
CN109493293A (en) A kind of image processing method and device, display equipment
Arsic et al. Improved lip detection algorithm based on region segmentation and edge detection
CN106446859B (en) Utilize the method for stain and the trace of blood in mobile phone front camera automatic identification human eye
CN109196517A (en) Comparison device and contrast method
JP2000348173A (en) Lip extraction method
Szlávik et al. Face analysis using CNN-UM
WO2005055144A1 (en) Person face jaw detection method, jaw detection system, and jaw detection program
KR19990026863A (en) Fuzzy Neuron Face Recognition Method
JPH07282260A (en) Recognizing method of individual by time series face image processing
WO2018116560A1 (en) Comparison device and comparison method
JP2001501750A (en) Eye localization filter

Legal Events

Date Code Title Description
A201 Request for examination
G160 Decision to publish patent application
E701 Decision to grant or registration of patent right
GRNT Written decision to grant
FPAY Annual fee payment

Payment date: 19990930

Year of fee payment: 4

LAPS Lapse due to unpaid annual fee