ANSI INCITS ISO IEC TR29794-5-2010 Information technology - Biometric sample quality - Part 5 Face image data《信息技术.生物统计样本质量.第5部分 表面图像数据》.pdf

《ANSI INCITS ISO IEC TR29794-5-2010 Information technology - Biometric sample quality - Part 5 Face image data《信息技术.生物统计样本质量.第5部分 表面图像数据》.pdf》由会员分享，可在线阅读，更多相关《ANSI INCITS ISO IEC TR29794-5-2010 Information technology - Biometric sample quality - Part 5 Face image data《信息技术.生物统计样本质量.第5部分 表面图像数据》.pdf（22页珍藏版）》请在麦多课文档分享上搜索。

1、 INCITS/ISO/IEC TR 29794-5-2010 Information technology Biometric sample quality Part 5: Face image data Copyright American National Standards Institute Provided by IHS under license with ANSI Not for ResaleNo reproduction or networking permitted without license from IHS-,-,-INCITS/ISO/IEC TR 29794-5

2、-2010 PDF disclaimer This PDF file may contain embedded typefaces. In accordance with Adobes licensing policy, this file may be printed or viewed but shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In downloading this

3、file, parties accept therein the responsibility of not infringing Adobes licensing policy. The ISO Central Secretariat accepts no liability in this area. Adobe is a trademark of Adobe Systems Incorporated. Details of the software products used to create this PDF file can be found in the General Info

4、 relative to the file; the PDF-creation parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.

5、Adopted by INCITS (InterNational Committee for Information Technology Standards) as an American National Standard. Date of ANSI Approval: 9/30/2010 Published by American National Standards Institute, 25 West 43rd Street, New York, New York 10036 Copyright 2010 by Information Technology Industry Coun

6、cil (ITI). All rights reserved. These materials are subject to copyright claims of International Standardization Organization (ISO), International Electrotechnical Commission (IEC), American National Standards Institute (ANSI), and Information Technology Industry Council (ITI). Not for resale. No pa

7、rt of this publication may be reproduced in any form, including an electronic retrieval system, without the prior written permission of ITI. All requests pertaining to this standard should be submitted to ITI, 1250 Eye Street NW, Washington, DC 20005. Printed in the United States of America ii ITIC

8、2010 All rights reserved Copyright American National Standards Institute Provided by IHS under license with ANSI Not for ResaleNo reproduction or networking permitted without license from IHS-,-,-INCITS/ISO/IEC TR 29794-5-2010 ITIC 2010 All rights reserved iiiContents Page Foreword iv Introduction.v

9、 1 Scope1 2 Normative references1 3 Terms and definitions .1 4 Abbreviated terms.1 5 Approaches to Face Image Quality .2 6 Categorization of Facial Quality.2 7 Facial Image Quality Analysis4 7.1 Dynamic Subject Characteristics 5 7.1.1 Subjects Behaviour5 7.1.2 Analysis Based on Statistical Differenc

10、es of the Left and Right Half of the Face5 7.2 Static Characteristics of the Acquisition Process .7 7.2.1 Image Resolution and Size.8 7.2.2 Noise.8 7.3 Characteristics of Image Acquisition8 7.3.1 Image Properties .8 7.3.2 Image Appearance.9 7.3.3 Illumination Intensity.9 7.3.4 Image Brightness 9 7.3

11、.5 Image Contrast 10 7.3.6 Exposure 11 7.3.7 Focus, Blur and Sharpness11 7.3.8 Colour .12 7.3.9 Subject-Camera distance .12 Bibliography13 Copyright American National Standards Institute Provided by IHS under license with ANSI Not for ResaleNo reproduction or networking permitted without license fro

12、m IHS-,-,-INCITS/ISO/IEC TR 29794-5-2010 iv ITIC 2010 All rights reservedForeword ISO (the International Organization for Standardization) and IEC (the International Electrotechnical Commission) form the specialized system for worldwide standardization. National bodies that are members of ISO or IEC

13、 participate in the development of International Standards through technical committees established by the respective organization to deal with particular fields of technical activity. ISO and IEC technical committees collaborate in fields of mutual interest. Other international organizations, gover

14、nmental and non-governmental, in liaison with ISO and IEC, also take part in the work. In the field of information technology, ISO and IEC have established a joint technical committee, ISO/IEC JTC 1. International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Pa

15、rt 2. The main task of the joint technical committee is to prepare International Standards. Draft International Standards adopted by the joint technical committee are circulated to national bodies for voting. Publication as an International Standard requires approval by at least 75 % of the national

16、 bodies casting a vote. In exceptional circumstances, the joint technical committee may propose the publication of a Technical Report of one of the following types: type 1, when the required support cannot be obtained for the publication of an International Standard, despite repeated efforts; type 2

17、, when the subject is still under technical devel opment or where for any other reason there is the future but not immediate possibility of an agreement on an International Standard; type 3, when the joint technical committee has coll ected data of a different kind from that which is normally publis

18、hed as an International Standard (“state of the art”, for example). Technical Reports of types 1 and 2 are subject to review within three years of publication, to decide whether they can be transformed into International Standards. Technical Reports of type 3 do not necessarily have to be reviewed u

19、ntil the data they provide are considered to be no longer valid or useful. Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights. ISO and IEC shall not be held responsible for identifying any or all such patent rights. ISO/IEC TR 29794-5

20、, which is a Technical Report of type 2, was prepared by Joint Technical Committee ISO/IEC JTC 1, Information technology, Subcommittee SC 37, Biometrics. ISO/IEC 29794 consists of the following parts, under the general title Information technology Biometric sample quality: Part 1: Framework Part 4:

21、Finger image data Technical Report Part 5: Face image data Technical Report Copyright American National Standards Institute Provided by IHS under license with ANSI Not for ResaleNo reproduction or networking permitted without license from IHS-,-,-INCITS/ISO/IEC TR 29794-5-2010 ITIC 2010 All rights r

22、eserved vIntroduction The purpose of this part of ISO/IEC 29794 is to define and specify methodologies for computation of objective, quantitative quality scores for facial images. Furthermore, the purpose, intent, and interpretation of face quality scores are defined. ISO/IEC 19794-5, Information te

23、chnology Biometric data interchange formats Part 5: Face image data, already gives some specifications that are related to scene constraints of the facial images, photographic properties of the facial images, and digital image attributes of the facial images. Within this part of ISO/IEC 29794, a sam

24、ple of a classification scheme of facial quality is exemplified and approaches for the determination of certain aspects of quality are introduced. Copyright American National Standards Institute Provided by IHS under license with ANSI Not for ResaleNo reproduction or networking permitted without lic

25、ense from IHS-,-,-Copyright American National Standards Institute Provided by IHS under license with ANSI Not for ResaleNo reproduction or networking permitted without license from IHS-,-,-TECHNICAL REPORT INCITS/ISO/IEC TR 29794-5-2010 ITIC 2010 All rights reserved 1Information technology Biometric

26、 sample quality Part 5: Face image data 1 Scope For aspects of quality specific to facial images, this part of ISO/IEC 29794: specifies terms and definitions that are useful in the specification, use and testing of face image quality metrics; defines the purpose, intent, and interp retation of face

27、image quality scores. Performance assessment of quality algorithms and standardization of quality algorithms are outside the scope of this part of ISO/IEC 29794. 2 Normative references The following referenced documents are indispensable for the application of this document. For dated references, on

28、ly the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies. ISO/IEC 29794-1, Information technology Biometric sample quality Part 1: Framework 3 Terms and definitions For the purposes of this document, the terms and definiti

29、ons given in ISO/IEC 29794-1 and the following apply. 3.1 comparison score numerical value (or set of values) resulting from a comparison 3.2 face quality assessment algorithm algorithm that computes a quality score for a given face image sample 3.3 facial image electronic image-based representation

30、 of the portrait of a person 4 Abbreviated terms CCD Charge-coupled device DCT Discrete Cosine Transform Copyright American National Standards Institute Provided by IHS under license with ANSI Not for ResaleNo reproduction or networking permitted without license from IHS-,-,-INCITS/ISO/IEC TR 29794-

31、5-2010 2 ITIC 2010 All rights reservedGCF Global Contrast Factor FQAA Face Quality Assessment Algorithm QS Quality Score FQS Face Quality Score QSN Quality Score Normalization 5 Approaches to Face Image Quality Face Image Quality can be defined in many ways, depending on the application. For the pur

32、pose of this part of ISO/IEC 29794 standard Face Image Quality is defined in relation to the use of facial images with automated face recognition systems. The performance of an automated face recognition system is affected by the amount of defect or the degree of imperfection present in the face ima

33、ge. The knowledge of quality can, and is currently being used to, process face images differently, by either invoking some image enhancement or normalization methods prior to feature extraction, invoking different matchers based on quality, or simply changing the threshold. The use of face image qua

34、lity assessment to enhance the overall performance of the system is increasing 3, 4, 5. A very important application of real-time quality analysis of faces is Face Recognition in Video, also referred to as Face in a Crowd, Recognition on the move, or Face at a Distance, e.g 21. This part of ISO/IEC

35、29794 shows some approaches for estimating Face Image Quality. The aim is to give the reader examples of assessment algorithms. Note, that these algorithms have pros and cons and no one algorithm is likely to be suitable for all facial images. Standardization of these algorithms is out of scope of t

36、his part of ISO/IEC 29794. The following related work is being done in ISO/IEC JTC1 SC37 1, 2: ISO/IEC 29794-1 suggests the use of Quality Algorithm Identification (QAID), or Quality Score Percentile Rank upon standardization of a Quality Score Normalization Dataset (QSND). This part of ISO/IEC 2979

37、4 adopts the following approach for face sample quality description: Specifying characterization of the facial quality and possible defects of face biometric samples in categorized aspects. Showing how FQAAs can be used to derive face quality scores (FQSs) related to specific characteristics and ass

38、ociated possible defects. An FQAA typically analyzes a face sample locally at the pixel or feature level and fuses the local analysis results over a global region. An FQS evaluates one or more characteristics and associated potential defects, and provides an indicator of the quality. A typical appro

39、ach of a system for generation of quality scores for facial images then takes the atomic FQSs generated by the FQAAs and combines them to a final quality score. The final quality score must predict performance metrics such as either false match or false non-match of an automatic facial image recogni

40、tion. 6 Categorization of Facial Quality Different factors affect the quality of the facial image with respect to biometric systems performance. A successful recognition will be based on the biometric characteristics of the subject and a number of factors that influence these characteristics such as

41、 variations (e.g. due to ageing) and the environmental conditions in the acquisition process: Influence of subjects characteristics on biometric performance, Influence of the acquisition process (including the capturing device) on biometric performance. Copyright American National Standards Institut

42、e Provided by IHS under license with ANSI Not for ResaleNo reproduction or networking permitted without license from IHS-,-,-INCITS/ISO/IEC TR 29794-5-2010 ITIC 2010 All rights reserved 3This classification is not sufficient, as it does not distinguish between static and dynamic characteristics and

43、properties: static subjects characteristics are related to anatomical characteristics of the subject, dynamic subject characteristics are related to subjects behaviour during the acquisition process, static properties of the acquisition process are related to physical properties of the capturing dev

44、ice and effects caused by the sample processing chain, dynamic properties of the acquisition process are related to environmental conditions during the capturing process. Table 1 shows a classification scheme that differentiates between the dynamic versus static properties as well as the subject ver

45、sus the acquisition process characteristics affecting facial quality. Table 1 Characterization of Facial Quality Subject characteristics Acquisition process Static Biological characteristics, like - anatomical characteristics (e.g. head dimensions, eye positions) - injuries and scars - ethnic group

46、- impairment Other static characteristics - Heavy facial wears, such as thick or dark glasses - Makeup - Permanent jewellery Acquisition process and capture device properties, like - image enhancement and data reduction process - physical properties (e.g. image resolution and contrast) - optical dis

47、tortions - static properties of the background, e.g. wallpaper - camera characteristics o sensor resolution - scene characteristics o geometric distortion Dynamic Subject characteristics and behaviour, like - closed eyes - (exaggerated) expression - hair across the eye - head pose - subject posing (

48、frontal / non frontal to camera) Scenery, like - dynamic characteristics of the background like moving objects - variation in lightning and related potential defects as o deviation from the symmetric lighting o uneven lighting on the face area o Extreme strong or weak illumination - subject posing , e.g. o too far (face too small), or too near (face too big) o out of focus (low sharpness) o partial occlusion of the face - Acquisition process and capture devic