JP2003216935A5 - - Google Patents

Download PDF

Info

Publication number
JP2003216935A5
JP2003216935A5 JP2002016550A JP2002016550A JP2003216935A5 JP 2003216935 A5 JP2003216935 A5 JP 2003216935A5 JP 2002016550 A JP2002016550 A JP 2002016550A JP 2002016550 A JP2002016550 A JP 2002016550A JP 2003216935 A5 JP2003216935 A5 JP 2003216935A5
Authority
JP
Japan
Prior art keywords
face
candidate
image processing
face area
area candidate
Prior art date
Legal status (The legal status 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 status listed.)
Withdrawn
Application number
JP2002016550A
Other languages
Japanese (ja)
Other versions
JP2003216935A (en
Filing date
Publication date
Application filed filed Critical
Priority to JP2002016550A priority Critical patent/JP2003216935A/en
Priority claimed from JP2002016550A external-priority patent/JP2003216935A/en
Publication of JP2003216935A publication Critical patent/JP2003216935A/en
Publication of JP2003216935A5 publication Critical patent/JP2003216935A5/ja
Withdrawn legal-status Critical Current

Links

Claims (12)

顔抽出の対象となる画像を入力する画像入力手段と、
入力画像に対し第１の画像処理を行う第１の画像処理手段と、
前記第１の画像処理により得られた特徴によって、入力画像から、顔を含むと思われる領域（以下、第１顔領域候補）を抽出する第１の顔抽出手段と、
入力画像に対し第２の画像処理を行う第２の画像処理手段と、
前記第２の画像処理により得られた特徴によって顔を含むと思われる領域（以下、第２顔領域候補）を抽出する第２の顔抽出手段と、
前記第１顔領域候補と前記第２顔領域候補とを統合し、統合顔領域候補を生成する統合手段とを備えたことを特徴とする顔抽出装置。An image input means for inputting an image to be subjected to face extraction;
First image processing means for performing first image processing on an input image;
A first face extracting means for extracting an area (hereinafter referred to as a first face area candidate) that is supposed to include a face from the input image according to the characteristics obtained by the first image processing;
Second image processing means for performing second image processing on the input image;
A second face extracting means for extracting an area (hereinafter referred to as a second face area candidate) that is supposed to include a face according to the feature obtained by the second image processing;
A face extraction apparatus comprising: an integration unit that integrates the first face area candidate and the second face area candidate to generate an integrated face area candidate. 入力された画像に対して画像処理を行って顔を抽出する顔抽出方法であって、
顔抽出の対象となる画像を入力するステップと、
入力画像に対し第１の画像処理を行うステップと、
前記第１の画像処理により得られた特徴によって、入力画像から、顔を含むと思われる領域（以下、第１顔領域候補）を抽出するステップと、
入力画像に対し第２の画像処理を行うステップと、
前記第２の画像処理により得られた特徴によって顔を含むと思われる領域（以下、第２顔領域候補）を抽出するステップと、
前記第１顔領域候補と前記第２顔領域候補とを統合し、統合顔領域候補を生成するステップとを備えたことを特徴とする顔抽出方法。A face extraction method for extracting a face by performing image processing on an input image,
Inputting an image for face extraction;
Performing first image processing on the input image;
Extracting a region (hereinafter referred to as a first face region candidate) that is supposed to include a face from the input image according to the characteristics obtained by the first image processing;
Performing second image processing on the input image;
Extracting an area (hereinafter referred to as a second face area candidate) that is supposed to include a face according to the feature obtained by the second image processing;
A face extraction method comprising: integrating the first face area candidate and the second face area candidate to generate an integrated face area candidate. 第１の画像処理は、顔の輪郭を抽出する処理であることを特徴とする請求項２記載の顔抽出方法。 3. The face extraction method according to claim 2 , wherein the first image processing is processing for extracting a face outline. 所定の同一候補判定条件に従って、第１顔領域候補と第２顔領域候補とが同一の検出対象に対する候補であるか否かを判定するステップを備え、同一候補判定を行った結果に基づいて第１顔領域候補と第２顔領域候補とを統合することを特徴とする請求項２または３記載の顔抽出方法。A step of determining whether or not the first face area candidate and the second face area candidate are candidates for the same detection target in accordance with a predetermined same candidate determination condition, and based on the result of performing the same candidate determination 1 face area candidate and claim 2 or 3 face extraction method, wherein the integrating and second face region candidates. 同一候補判定条件として、第１顔領域情報の領域内に第２顔領域情報の重心が存在するか否かを含むことを特徴とする請求項４記載の顔抽出方法。5. The face extraction method according to claim 4 , wherein the same candidate determination condition includes whether or not the center of gravity of the second face area information exists in the area of the first face area information. 同一候補判定条件として、第１顔領域候補の内部に設定された同一判定領域内に、第２顔領域候補の重心が存在するか否かを含むことを特徴とする請求項４記載の顔抽出方法。5. The face extraction according to claim 4 , wherein the same candidate determination condition includes whether or not the center of gravity of the second face area candidate exists within the same determination area set inside the first face area candidate. Method. 同一候補判定条件として、第１顔領域候補にサイズ比較係数を乗算した比較サイズと第２顔領域候補の領域サイズとの差分が閾値以内であるか否かを含むことを特徴とする請求項４〜６の何れかに記載の顔抽出方法。As the same candidate determination condition, claims the difference between the comparative size obtained by multiplying the size comparison coefficients to the first face region candidate area size of the second face region candidate, characterized in that it comprises whether within the threshold 4 The face extraction method according to any one of -6 . 同一候補か否かの判定結果に基づいて、第１顔領域候補に対する信頼度と第２顔領域候補に対する信頼度とから、統合顔領域候補に対する統合結果信頼度を計算するステップを備えたことを特徴とする請求項４〜７の何れかに記載の顔抽出方法。A step of calculating an integrated result reliability for the integrated face area candidate from the reliability for the first face area candidate and the reliability for the second face area candidate based on the determination result of whether or not they are the same candidate; The face extraction method according to any one of claims 4 to 7 , characterized in that: 統合結果信頼度を計算するステップでは、同一候補と判定された場合には、第１顔領域候補の信頼度と第２顔領域候補の信頼度とを、予め設定された比率で加算して得られる値を、統合顔領域候補に対する統合結果信頼度とすることを特徴とする請求項８記載の顔抽出方法。In the step of calculating the integration result reliability, when it is determined that the candidates are the same, the reliability of the first face area candidate and the reliability of the second face area candidate are added at a preset ratio. 9. The face extraction method according to claim 8 , wherein the obtained value is an integration result reliability for the integrated face area candidate. 統合顔領域候補を生成するステップでは、第１顔領域候補と第２顔領域候補の信頼度の比率により統合顔領域候補の位置、大きさを計算することを特徴とする請求項８または９記載の顔抽出方法。In the step of generating an integrated face region candidate, the first face region candidate and the position of the integrated face region candidates by the ratio of the reliability of the second face region candidate, claim 8 or 9, wherein the calculating the magnitude Face extraction method. 統合顔領域候補を生成するステップでは、同一候補と判定された第１顔領域候補と第２顔領域候補のうち、信頼度が最大となる候補によって前記統合顔領域候補を求め、信頼度が最大とならない候補を削除することを特徴とする請求項８〜１０の何れかに記載の顔抽出方法。In the step of generating the integrated face area candidate, the integrated face area candidate is obtained by a candidate having the maximum reliability among the first face area candidate and the second face area candidate determined to be the same candidate, and the reliability is maximized. The face extraction method according to claim 8 , wherein candidates that do not become are deleted. 入力された画像に対して画像処理を行って顔を抽出する顔抽出プログラムであって、
入力画像に対し第１の画像処理を行うステップと、
前記第１の画像処理により得られた特徴によって、入力画像から、顔を含むと思われる領域（以下、第１顔領域候補）を抽出するステップと、
入力画像に対し第２の画像処理を行うステップと、
前記第２の画像処理により得られた特徴によって顔を含むと思われる領域（以下、第２顔領域候補）を抽出するステップと、
前記第１顔領域候補と前記第２顔領域候補とを統合し、統合顔領域候補を生成するステップとを備えたことを特徴とする顔抽出プログラムを記録した記録媒体。A face extraction program that performs image processing on an input image and extracts a face,
Performing first image processing on the input image;
Extracting a region (hereinafter referred to as a first face region candidate) that is supposed to include a face from the input image according to the characteristics obtained by the first image processing;
Performing second image processing on the input image;
Extracting an area (hereinafter referred to as a second face area candidate) that is supposed to include a face according to the feature obtained by the second image processing;
A recording medium on which a face extraction program is recorded, comprising the step of integrating the first face area candidate and the second face area candidate to generate an integrated face area candidate.
JP2002016550A 2002-01-25 2002-01-25 Face extracting device and method, and storage medium with those stored therein Withdrawn JP2003216935A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2002016550A JP2003216935A (en) 2002-01-25 2002-01-25 Face extracting device and method, and storage medium with those stored therein

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2002016550A JP2003216935A (en) 2002-01-25 2002-01-25 Face extracting device and method, and storage medium with those stored therein

Publications (2)

Publication Number Publication Date
JP2003216935A JP2003216935A (en) 2003-07-31
JP2003216935A5 true JP2003216935A5 (en) 2005-07-21

Family

ID=27652579

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2002016550A Withdrawn JP2003216935A (en) 2002-01-25 2002-01-25 Face extracting device and method, and storage medium with those stored therein

Country Status (1)

Country Link
JP (1) JP2003216935A (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2395781A (en) * 2002-11-29 2004-06-02 Sony Uk Ltd Face detection
JP4534571B2 (en) * 2004-04-22 2010-09-01 三菱電機株式会社 Face detection device
WO2006025185A1 (en) * 2004-08-31 2006-03-09 Matsushita Electric Industrial Co., Ltd. Monitoring recorder and its method
JP2006215685A (en) 2005-02-02 2006-08-17 Funai Electric Co Ltd Photo printer
JP4431532B2 (en) 2005-09-16 2010-03-17 富士フイルム株式会社 Target image position detecting device and method, and program for controlling target image position detecting device
JP2009238202A (en) * 2008-03-03 2009-10-15 Dainippon Screen Mfg Co Ltd Contour extraction device, contour extraction method and contour extraction program
JP6261206B2 (en) * 2013-06-28 2018-01-17 キヤノン株式会社 Information processing apparatus, information processing method, and program

Similar Documents

Publication Publication Date Title
JP2008271268A5 (en)
TW200511387A (en) Template creation method and device, pattern detection method, position detection method and device, exposure method and device, device manufacturing method, and recording medium to record the template creation program
JP2009053815A (en) Subject tracking program and subject tracking device
EP1418583A3 (en) Data recording device, data recording method, program and data recording medium
JP2005157569A5 (en)
CN110827249A (en) Electronic equipment backboard appearance flaw detection method and equipment
US8938132B2 (en) Image collation system, image collation method and computer program
CN110827246A (en) Electronic equipment frame appearance flaw detection method and equipment
JP2003216935A5 (en)
JPWO2018235219A1 (en) Self-location estimation method, self-location estimation device, and self-location estimation program
JP2006244074A (en) Moving object close-up frame detection method and program, storage medium storing program, moving object close-up shot detection method, moving object close-up frame or shot detection method and program, and storage medium storing program
JP5192437B2 (en) Object region detection apparatus, object region detection method, and object region detection program
CN111310912A (en) Machine learning system, domain conversion device, and machine learning method
EP1465380A8 (en) Device which executes authentication processing by using offline information, and device authentication method
Wang et al. Multimodal and multi-task audio-visual vehicle detection and classification
JP2007140695A (en) Suspicious face detection system, suspicious face detection method and suspicious face detection program
KR101908938B1 (en) Detection of foreground object using background model-based temporal minimum filtering and log histogram
WO2016142965A1 (en) Video processing device, video processing method, and storage medium storing video processing program
JP2006338368A5 (en)
JP2008085539A (en) Program, detection method, detector, image processing method and image processor
JP2003115992A5 (en)
WO2022102038A1 (en) Information processing device, generation method, and generation program
JP6360674B2 (en) Evaluation device for voids in solder and evaluation method for voids in solder
KR102093208B1 (en) Character recognition device based on pixel analysis and operating method thereof
JP6725185B2 (en) Acoustic signal separation device and acoustic signal separation method