TW201035877A - Human face scoring method - Google Patents

Human face scoring method Download PDF

Info

Publication number
TW201035877A
TW201035877A TW98109518A TW98109518A TW201035877A TW 201035877 A TW201035877 A TW 201035877A TW 98109518 A TW98109518 A TW 98109518A TW 98109518 A TW98109518 A TW 98109518A TW 201035877 A TW201035877 A TW 201035877A
Authority
TW
Taiwan
Prior art keywords
face
image
nose
scoring
golden ratio
Prior art date
Application number
TW98109518A
Other languages
Chinese (zh)
Inventor
qi-rong Weng
Ming-Song Xiao
Original Assignee
Altek Corp
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 Altek Corp filed Critical Altek Corp
Priority to TW98109518A priority Critical patent/TW201035877A/en
Publication of TW201035877A publication Critical patent/TW201035877A/en

Links

Landscapes

  • Image Analysis (AREA)
  • Studio Devices (AREA)

Abstract

This invention discloses a human face scoring function, which comprises obtaining an image; analyzing whether the image contains a human face; then using a face recognition method to cut the image into a plurality of sections for detecting the human face in the image, so as to obtain the characteristic value of the human face; then using the golden ratio of human face as the scoring standard to score the characteristic value of the human face; and finally outputting the message of scoring result. The scoring function can be installed in any multimedia electronic product capable of capturing images to provide users with new amusement during photograph taking.

Description

201035877 六、發明說明: 【發明所屬之技術領域】 本發明係有關-種人麟識之制,_是_—種人齡攝後可對 其臉部作評價之功能技術。 【先前技術】 • 人臉是千變換化,雖然固定由五官之排列組合,人臉基本架構上係由 眉毛、眼睛、鼻子、嘴巴以及頭部兩側邊的一對耳朵,若缺乏一者則無法 成就-張漂亮的人臉,而隨著不-樣的基因遺傳,使其五官之分佈結構及 〇 其形狀造就千千萬萬種人臉’因此,業麵由人賴識程式將人臉特徵歸 納分類,做其應用。 而’人臉辨識技術已十分普^應用於各個角落,例如,根據中華民國 公告號第⑽撕號專利揭示—種保全系統中辨識真實人臉之方*,保全人 貞或警謂監視騎機巾可疑人士峨孔’藉由人賴難術_其特徵 長相以追拿嫌犯的憑據,或刑事訴訟時作為控告之證據;根據中華民國 公告號第M317045號專利所揭示應用於電腦、手機或其他行動裝置中,人 G麟識技娜為_登功取控制前之關鍵轉,為制者之保密資料多 伤安全保障措施;再根據中華民國公開號第2〇〇8〇2132號之申請專利, 其提供-種臉部分類方法,及其分酿紀錄之程式,美容業者可利用相關 功能將人臉作分類,以數位化臉部資訊,方便對消費者的作臉部的美容建 議。 因此,近來人臉辨識技術多應用於如上述之保全、個人資料加密或是 美合業等領域’但從未有業者未將人臉辨識技術與評價人臉美麗值作結 3 201035877 合,再者’如能將此種美麗值評分技術應用於照相機 子多媒艘產品上,則可再提供消費者一項拍攝人臉之樂趣,更可 愛美的族群消費購買。 基於上述的缺失,本剌係提出—種人臉評分方法以提供市場另一 - 種應用人臉辨識技術之領域。 【發明内容】 本發明之主要目的,係提出—種人麟分舰,使柿可增加一 評價人臉分數的魏,以提供大眾拍照之樂趣。 Ο 為達上述之目的,本發明提供一種人臉評分方法,係先取得-影像; 辨別此影像是否具有人臉,並將具有人臉之影像輸出;接收具有人臉之影 像’且利用臉部職法偵測該人臉,並得此影像中人臉之特徵值;以人臉 黃金比例條件為標準’對人臉之特徵值進行評分;最後輸出評分結果之訊 息。 。 〇 底下藉由频實施繼麵_圖式詳加朗,料容_解本發明 之目的、技術内容、特點及其所達成之功效。 【實施方式】 就人類愛美的天鶴言,尤其時下年輕族群非倾愛用手機或相機自 拍’將自身年輕美麗的一面拍攝下_為留念,而本發明提供一種人臉評 分功能’可配合攝取人郞像,將其人臉建議評分,則可以增添拍攝時的 樂趣。 在此’本發糊示-種人臉評分方法,請參見第1圓係為本發明之 步驟流程圖’如步驟S10所示係先取得一影像,係利用影像掏取器,例如 201035877 相機、攝影機或附有CCD鏡頭之電子多媒體產品或影像劉覽器,例如數位 相框;如步驟S12所示,辨別所取得之影像是否為具有人臉之影像;如步 驟S16所示,接收經過判斷具有人臉之影像,且利用臉部辨識法偵測影像 中的人臉,並得其此人臉特徵值;如步驟S18所示’提供一評分標準對此 人臉的特徵值進行評分;最後輸出評分結果的訊息,如步驟S2〇所示。 料’臉雜識法係购鼠影像,係將影像切赋複魅域方式以 _此影像,而且用上述臉部辨識法方式此種臉部辨識法可辨識區分此人 〇 臉之性別。並且’評分標準係以人臉黃金比例條件作為最美麗條件之指標, 最後’該評分結果係可以數值或文字描述呈現。 人臉黃金比例條件如下: 1. 參見第2a圖所示,其綱至眉毛a、眉毛至鼻子b與鼻子至下巴c 之比例符合1 : 1 : 1。 2. 參見第2b圖所示’眼睛長度d所佔整臉寬度e之比例為一眼長度心 3. 參見第2b圖所示,雙眼之距離f為一眼長度d。 〇 4.參見第215_示,眉毛至鼻子g之雜鱗於耳朵長度。 5. 參見第2b圖所示,眼睛長度d為臉部寬度e的五分之一。 6. 參見第2C圖所示,鼻子長度h約為中臉部長度i的67%,其中, 料長麟從鼻根輕鼻頭,愤部長度魏觸至鼻翼。 7·鼻頭的高度為鼻子長度的67%。 8. 眉毛的位置男性則位於眉弓骨上。 9. 鼻頭與人中之間的傾斜角度,男性的傾斜角度為100度至1〇3度。 10. 以鼻根部與下巴下拉長成一垂直線,鼻子隆起的高度與其形成的角 5 201035877 度,男性的角度為36度。 U•眉毛的位置女性約在於眉弓骨上1公分。 12. 鼻頭與人中之間的傾斜角度,女性的傾斜角度為⑽度至⑽度。 13. 以鼻根部與下巴下拉魏__垂直線,鼻子隆起的高度與其形成的角 度’女性的角度為34度。 «上為人臉黃金_之條件,本發狀人臉評㈣法雜由這組條件 作為最美麗人臉之指標,經由人臉辨識法取得待測人臉影像之特徵值,將 〇 碰雌人臉黃姐·件_運雜,即可制姆於此黃金比例條件 之評分結果。 以上所述係藉由實施例說明本發明之特點,其目的在使熟習該技術者 能暸解本發明之内容並據以實施,而非限定本發明之專利範圍,故,凡其 他未脫離本發明所揭示之精神所完成之等效修飾或修改,仍應包含在以下 所述之申請專利範圍中。 【圖式簡單說明】 ◎ S1®為本發明之本發明之流程方塊圖。 第2这〜2c圖為人臉黃金比例之示意圖。 【主要元件符號說明】201035877 VI. Description of the invention: [Technical field to which the invention pertains] The present invention relates to a system for cultivating a person's knowledge, and _ is a functional technique for evaluating the face of a person after being photographed. [Prior Art] • The face is a thousand-transformation. Although it is fixed by a combination of facial features, the basic structure of the face is composed of an eyebrow, an eye, a nose, a mouth, and a pair of ears on both sides of the head. If one is lacking, Unable to achieve - a beautiful face, and with the genetic inheritance, the distribution structure of the five senses and its shape create thousands of faces. Feature induction classification, do its application. And 'face recognition technology has been applied to every corner, for example, according to the Republic of China Announcement No. (10) tear-off patent disclosure - a kind of real-life system to identify the real face *, security guards or warnings to monitor the ride Suspicious person's pupil 'by the person's refusal _ its characteristics look to chase the suspect's evidence, or evidence of the accusation in criminal proceedings; according to the Republic of China Bulletin No. M317045 patent disclosed in computers, mobile phones or other In the mobile device, the person G is aware of the key turn before the control, and the security information for the confidential information of the system is injured; and then the patent application according to the Republic of China Public No. 2〇〇8〇2132 It provides a method of face-type methods and a program for dividing the records. The beauty industry can use the relevant functions to classify the faces to digitize facial information and facilitate facial cosmetic advice for consumers. Therefore, face recognition technology has recently been applied to such areas as security, personal data encryption, or the United States. However, no one has ever used face recognition technology to evaluate the face value of the face. 201038877 'If you can apply this beautiful value scoring technology to the camera multi-media ship products, you can provide consumers with a fun to shoot people's faces, more lovely and beautiful people to buy. Based on the above-mentioned shortcomings, the present invention proposes a face scoring method to provide another field of application of face recognition technology. SUMMARY OF THE INVENTION The main object of the present invention is to propose a kind of human-individual ship, so that the persimmon can add a Wei to evaluate the face score to provide the fun of the public photograph. Ο For the above purpose, the present invention provides a face scoring method, which first acquires an image; discriminates whether the image has a human face, and outputs an image with a human face; receives an image with a human face and utilizes the face The job method detects the face and obtains the feature value of the face in the image; the face value of the face is used as a standard to score the feature value of the face; finally, the message of the score result is output. . 〇 Under the circumstance, the implementation of the continuation of the _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ [Embodiment] In the case of the human beings who love beauty, especially the young people who are not in love with mobile phones or camera self-portraits, take their own young and beautiful side to take pictures, and the present invention provides a face-scoring function that can be combined with ingestion. People's portraits, with their face suggestions, can add fun when shooting. In the present invention, the first round is a flow chart of the steps of the present invention. As shown in step S10, an image is first acquired, and an image capture device such as a 201035877 camera is used. a camera or an electronic multimedia product or image viewer with a CCD lens, such as a digital photo frame; as shown in step S12, it is determined whether the acquired image is an image with a human face; as shown in step S16, the receiving is judged to have a human The image of the face, and the face recognition method is used to detect the face in the image, and the face feature value is obtained; as shown in step S18, 'a score criterion is provided to score the feature value of the face; and finally the score is output. The resulting message is as shown in step S2. The face-familiar method is to purchase the mouse image, and the image is cut into the enchantment mode to use the image, and the face recognition method can recognize the gender of the person's face. And 'the scoring standard is based on the face golden ratio condition as the most beautiful conditional indicator, and finally the result of the rating can be presented in numerical or textual description. The golden ratio of face is as follows: 1. See Figure 2a, the ratio of eyebrow a to eyebrow to nose b to nose to chin c is 1: 1 : 1. 2. See Figure 2b for the 'eye length d' of the entire face width e as the length of one eye. 3. See Figure 2b, the distance f between the eyes is the length d of one eye. 〇 4. See 215_, the size of the eyebrows to the nose g is in the length of the ear. 5. Referring to Figure 2b, the eye length d is one-fifth of the width e of the face. 6. Referring to Figure 2C, the length h of the nose is about 67% of the length i of the middle face. Among them, the long stalk from the nose is lightly nosed, and the length of the anger is touching the nose. 7. The height of the nose is 67% of the length of the nose. 8. The position of the eyebrows is located on the eyebrow bone. 9. The angle of inclination between the nose and the person, the angle of inclination of the male is 100 degrees to 1〇3 degrees. 10. With the base of the nose and the chin pulled down into a vertical line, the height of the nose bulges and the angle formed by it 5 201035877 degrees, the angle of the male is 36 degrees. U• The position of the eyebrows is about 1 cm on the eyebrow bone. 12. The angle of inclination between the nose and the person, the angle of inclination of the woman is (10) degrees to (10) degrees. 13. With the nose root and chin pulled down the Wei__ vertical line, the height of the nose bulge and the angle formed by it's angle of 34 degrees. «The condition of the face is gold _, the hair face evaluation (four) method is used as the most beautiful face indicator, and the feature value of the face image to be tested is obtained by the face recognition method, and the female face will be touched. Face yellow sister · pieces _ Yun miscellaneous, you can make the score of this golden ratio condition. The above description of the embodiments of the present invention is intended to be understood by those skilled in the art, and the invention may be practiced without departing from the scope of the invention. Equivalent modifications or modifications made by the spirit of the invention should still be included in the scope of the claims described below. BRIEF DESCRIPTION OF THE DRAWINGS ◎ S1® is a block diagram of the present invention. The second figure ~ 2c is a schematic diagram of the golden ratio of the face. [Main component symbol description]

Claims (1)

201035877 七、申請專利範圍: 1. 一種人臉評分方法,其步驟包含: 取得一影像; 判斷該影像是否為具人臉之影像,並將判斷為具人臉之影像輸出 以得該影像 接收經判斷為具人臉之影像並利用臉部辨識法俄測該影像, 中人臉之特徵值; 提供一評分標準以對該人臉之特徵值進行評分;以及 輸出評分結果之訊息。 D 如申請專利範圍第!項所述之人臉評分方法,其中該臉部辨識法包括以 切割該影像為複數區域之方式偵測該影像中之人臉。 3. 如申請專利範M i項所述之人臉評分方法’射該臉部辨識法包括區 分該影像中人臉之性別。 4. 如申請專利第丨項所述之人臉評分方法,其中評分標準係以人臉黃 金比例條件作為指標。 ❹5.如”專利範圍f 4項所述之人臉評分方法,其中該人臉黃金比例條件 係為該人臉影像之額頭至眉毛、該眉毛至鼻子'該鼻子至下巴之比例係 6. 如申請專利範圍第4項所述之人臉評分方法,該人臉黃金比例條件更包 括該人臉影像之一眼睛所佔人臉之比例係為該眼睛長度。 7. 如申清專利範圍第4項所述之人臉評分方法,該人臉黃金比例條件更包 括該人臉影像之眼睛與另一眼睛距離係為該眼睛長度。 8·如申請專利範Μ 4項所述之人断分方法,該人臉黃金比例條件更包 201035877 括該眼睛長度為該人臉寬度的五分之一。 9. 如申請專利範圍第4項所述之人臉評分方法,該人臉黃金比例條件更包 括該人臉影像之眉毛至鼻子之距離係為耳朵長度。 10. 如申請專利範圍第4項所述之人臉評分方法,該人臉黃金比例條件更包 括該人臉影像之鼻子長度約為中臉部長度之67%。 11. 如申請專利範圍第10項所述之人臉評分方法,其中該鼻子長度為該人臉 影像之鼻根部至鼻頭兩者之距離。 12. 如申請專利範圍第10項所述之人臉評分方法,其中該中臉部長度為該人 臉影像之眉間至鼻翼兩者之距離。 13. 如申請專利範圍第4項所述之人臉評分方法,該人臉黃金比例條件更包 括該人臉影像之鼻頭之商度為鼻子長度的67%。 14. 如申請專利範圍第4項所述之人臉評分方法,該人臉黃金比例條件更包 括該人臉影像之鼻根部寬度與鼻翼寬度等長。 15. 如申請專利範圍第4項所述之人臉評分方法,其中該人臉黃金比例條件 區分為男性黃金比例條件與女性黃金比例條件。 16. 如申請專利範圍第15項所述之人臉評分方法,其中該男性黃金比例條件 係為該人臉影像之眉毛的位置係位於該眉弓骨上。 Π.如申請專利範圍第I6項所述之人臉評分方法,該男性黃金比例條件更包 括該人臉影像之鼻頭與人中兩者之傾斜角度係為1〇〇度至度。 18.如申請賴_第I6酬述之人臉評分方法,該雜黃金_條件更包 括以該人臉影像之鼻子根部和下巴所形成一垂直線與鼻子隆起角度係為 36度。 8 201035877 19. 如申請專利範園第15項所述之人臉評分方法,其中該女性黃金比例條件 係為該人臉影像之眉毛之位置係位於眉弓骨上方一公分。 20. 如申請專利範圍第16項所述之人臉評分方法,該女性黃金比例條件更包 括該人臉影像之鼻頭與人中之間的傾斜角度係為1〇5度至1〇8度。 21. 如申請專利範圍第16項所述之人臉評分方法,該女性黃金比例條件更包 触該人臉影像之鼻子根部和下巴卿成—垂直線無鼻子隆起角度係 為34度。 〇 22.如申請專利範圍第1項所述之人臉評分方法,其中該評分結果係可以數 值或文字描述呈現。 二:::广圍第1項所述之人臉評分方法’其中係利用-影像操取器 子多媒體產。… 彳機攝-機 '附有CCD鏡頭之電 7干11座〇σ、或影像瀏覽器。201035877 VII. Patent application scope: 1. A face scoring method, the steps comprising: obtaining an image; determining whether the image is an image with a human face, and outputting the image as a human face to obtain the image receiving Judging as an image with a human face and using the face recognition method to measure the feature value of the face in the image; providing a scoring standard to score the feature value of the face; and outputting the message of the scoring result. D If you apply for a patent scope! The face scoring method of the item, wherein the face recognition method comprises detecting a face in the image by cutting the image into a plurality of regions. 3. The face scoring method as described in the application patent specification M i, the method of detecting the face includes distinguishing the gender of the face in the image. 4. The face scoring method described in the patent application, wherein the scoring standard is based on the proportion of face gold. ❹ 5. The method of face grading according to claim 4, wherein the face ratio of the face is the forehead of the face image to the eyebrow, the eyebrow to the nose, the ratio of the nose to the chin is 6. In the face scoring method described in claim 4, the face golden ratio condition further includes the proportion of the face occupied by the eye of the face image as the length of the eye. 7. In the face scoring method described in the item, the face golden ratio condition further includes the distance between the eye of the face image and the other eye is the length of the eye. 8. The method for dividing a person as described in the patent application 4 The face golden ratio condition is further included in 201035877, and the length of the eye is one-fifth of the width of the face. 9. The face-grading method described in claim 4 of the patent scope, the face golden ratio condition further includes The distance from the eyebrow to the nose of the face image is the length of the ear. 10. According to the face scoring method described in claim 4, the face golden ratio condition further includes the length of the nose of the face image is about medium Face length The face scoring method according to claim 10, wherein the nose length is the distance from the nose root to the nose of the face image. 12. As described in claim 10 The face scoring method, wherein the middle face length is the distance from the eyebrow of the face image to the nose wing. 13. The face scale method according to item 4 of the patent application scope, the face golden ratio condition is more The degree of the nose including the face image is 67% of the length of the nose. 14. The face scale method according to item 4 of the patent application, the face golden ratio condition further includes the nose root width of the face image. The length of the nose is as long as 15. The face scoring method described in claim 4, wherein the face golden ratio condition is divided into a male golden ratio condition and a female golden proportional condition. The method for scoring a face according to the item, wherein the male golden ratio condition is that the position of the eyebrow of the facial image is located on the eyebrow arch. Π. The face scoring method described in claim 1 of the patent scope, The male golden ratio condition further includes the inclination angle of the nose and the person of the face image is 1 degree to degree. 18. If the application is based on the face evaluation method of the I6 reward, the mixed gold _ The condition further includes a vertical line formed by the nose root and the chin of the face image and a nose bulging angle of 36 degrees. 8 201035877 19. The face scoring method described in claim 15 of the patent application, wherein the female The golden ratio condition is that the position of the eyebrow of the face image is one centimeter above the eyebrow bone. 20. The face scale method described in claim 16 of the patent application, the female golden ratio condition further includes the face image. The angle of inclination between the nose and the person is from 1 〇 5 degrees to 1 〇 8 degrees. 21. For the face scoring method described in claim 16 of the patent application, the female golden ratio condition further covers the nose root of the face image and the chin Qingcheng-vertical line without the nose bulge angle of 34 degrees. 〇 22. The face scoring method of claim 1, wherein the scoring result is presented in a numerical or textual description. 2::: The face scoring method described in item 1 of Guangwei ’ is the use of the image manipulation device. ... 彳机摄-机 'With CCD lens power 7 dry 11 〇 σ, or image viewer.
TW98109518A 2009-03-24 2009-03-24 Human face scoring method TW201035877A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW98109518A TW201035877A (en) 2009-03-24 2009-03-24 Human face scoring method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW98109518A TW201035877A (en) 2009-03-24 2009-03-24 Human face scoring method

Publications (1)

Publication Number Publication Date
TW201035877A true TW201035877A (en) 2010-10-01

Family

ID=44856028

Family Applications (1)

Application Number Title Priority Date Filing Date
TW98109518A TW201035877A (en) 2009-03-24 2009-03-24 Human face scoring method

Country Status (1)

Country Link
TW (1) TW201035877A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110990607A (en) * 2019-11-25 2020-04-10 成都市喜爱科技有限公司 Game photo screening method, device, server and computer-readable storage medium
TWI731447B (en) * 2018-12-03 2021-06-21 日商碧柏熙股份有限公司 Beauty promotion device, beauty promotion system, beauty promotion method, and beauty promotion program

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI731447B (en) * 2018-12-03 2021-06-21 日商碧柏熙股份有限公司 Beauty promotion device, beauty promotion system, beauty promotion method, and beauty promotion program
CN110990607A (en) * 2019-11-25 2020-04-10 成都市喜爱科技有限公司 Game photo screening method, device, server and computer-readable storage medium
CN110990607B (en) * 2019-11-25 2023-07-18 成都市喜爱科技有限公司 Method, apparatus, server and computer readable storage medium for screening game photos

Similar Documents

Publication Publication Date Title
JP5190560B2 (en) Content output apparatus, content output method, content output program, and recording medium on which content output program is recorded
US7319780B2 (en) Imaging method and system for health monitoring and personal security
Alharbi et al. To mask or not to mask? balancing privacy with visual confirmation utility in activity-oriented wearable cameras
WO2015122195A1 (en) Impression analysis device, game device, health management device, advertising support device, impression analysis system, impression analysis method, program, and program recording medium
US20150199730A1 (en) Sentiments based transaction systems and methods
CA3050456C (en) Facial modelling and matching systems and methods
CN111902821A (en) Detecting motion to prevent recognition
CN101859369A (en) Face partition method
Shein Filtering for beauty
Gill Perfect: Feeling judged on social media
TW201035877A (en) Human face scoring method
Bonner et al. When filters escape the smartphone: Exploring acceptance and concerns regarding augmented expression of social identity for everyday AR
JP4937639B2 (en) Authentication method, authentication device, authentication program, and recording medium on which the program is recorded
Bouma et al. Measuring cues for stand-off deception detection based on full-body nonverbal features in body-worn cameras
Sakthivel et al. Online Education Pedagogy Approach
Malepati et al. Extraction of human facial behavioural expressions using Haar like features compared with Haar cascade classifier
Hughes et al. Sex differences in the assessment of pain versus sexual pleasure facial expressions.
CN112668833A (en) Staff work arrangement method, device, equipment and medium based on artificial intelligence
JPWO2020116069A1 (en) Information processing equipment and programs
JP7355969B1 (en) Information processing system, information processing method, program
CN112309018B (en) Image display method and device, computer equipment and storage medium
KR102630872B1 (en) Apparatus and method for learning facial expression recognition
EP4170609A1 (en) Automated filter selection for altering a stream
CN117853977B (en) Method, system, equipment and medium for intelligent evaluation through video
Oh et al. A study on user-experience evaluation of VR contents using questionnaire