ANSI INCITS 385-2004 Information Technology - Face Recognition Format for Data Interchange.pdf

1、American National StandardDeveloped byfor Information Technology Face Recognition Formatfor Data InterchangeANSI INCITS 385-2004 (R2009)ANSIINCITS385-2004ANSIINCITS 385-2004(R2009)American National Standardfor Information Technology Face Recognition Formatfor Data InterchangeSecretariatInformation T

2、echnology Industry CouncilApproved May 13, 2004American National Standards Institute, Inc.AbstractThis standard specifies definitions of photographic (environment, subject pose, focus, etc.) properties, dig-ital image attributes and a face interchange format for relevant applications, including huma

3、n examinationand computer automated face recognition.Approval of an American National Standard requires review by ANSI that therequirements for due process, consensus, and other criteria for approval havebeen met by the standards developer.Consensus is established when, in the judgement of the ANSI

4、Board ofStandards Review, substantial agreement has been reached by directly andmaterially affected interests. Substantial agreement means much more thana simple majority, but not necessarily unanimity. Consensus requires that allviews and objections be considered, and that a concerted effort be mad

5、etowards their resolution.The use of American National Standards is completely voluntary; theirexistence does not in any respect preclude anyone, whether he has approvedthe standards or not, from manufacturing, marketing, purchasing, or usingproducts, processes, or procedures not conforming to the s

6、tandards.The American National Standards Institute does not develop standards andwill in no circumstances give an interpretation of any American NationalStandard. Moreover, no person shall have the right or authority to issue aninterpretation of an American National Standard in the name of the Ameri

7、canNational Standards Institute. Requests for interpretations should beaddressed to the secretariat or sponsor whose name appears on the titlepage of this standard.CAUTION NOTICE: This American National Standard may be revised orwithdrawn at any time. The procedures of the American National Standard

8、sInstitute require that action be taken periodically to reaffirm, revise, orwithdraw this standard. Purchasers of American National Standards mayreceive current information on all standards by calling or writing the AmericanNational Standards Institute.American National StandardPublished byAmerican

9、National Standards Institute, Inc.25 West 43rd Street, New York, NY 10036Copyright 2004 by Information Technology Industry Council (ITI)All rights reserved.No part of this publication may be reproduced in anyform, in an electronic retrieval system or otherwise,without prior written permission of ITI

10、, 1250 Eye Street NW, Washington, DC 20005. Printed in the United States of AmericaCAUTION: The developers of this standard have requested that holders of patents that may berequired for the implementation of the standard disclose such patents to the publisher. However,neither the developers nor the

11、 publisher have undertaken a patent search in order to identifywhich, if any, patents may apply to this standard. As of the date of publication of this standardand following calls for the identification of patents that may be required for the implementation ofthe standard, no such claims have been m

12、ade. No further patent search is conducted by the de-veloper or publisher in respect to any standard it processes. No representation is made or impliedthat licenses are not required to avoid infringement in the use of this standard.iContentsPageForeword vIntroduction .ixDocument Overview 11 Scope. 4

13、2 Conformance . 43 Normative References . 44 Terms and Definitions 55 The Face Record Format. 65.1 Overview 65.2 Data Conventions 85.2.1 Byte Ordering. 85.2.2 Numeric values 85.3 The CBEFF Header . 85.4 The Facial Header Block 95.4.1 Format identifier. 95.4.2 Version Number. 95.4.3 Record length. 95

14、.4.4 Number of Facial Images. 95.5 The Facial Information Block . 105.5.1 Block Length 105.5.2 Number of Feature Points 105.5.3 Gender. 105.5.4 Eye Color . 105.5.5 Hair Color. 115.5.6 Feature Mask. 135.5.7 Expression . 145.5.8 Pose Angles. 155.5.9 Pose Angle Uncertainty . 175.6 The Feature Block 185

15、.6.1 MPEG4 Features . 185.6.2 Center of Facial Features 195.6.3 The Facial Feature Block Encoding. 205.7 The Image Information Block. 215.7.1 Facial Image Type . 215.7.2 Image Data Type . 225.7.3 Width 225.7.4 Height. 225.7.5 Image Color Space 235.7.6 Source Type. 245.7.7 Device Type. 245.7.8 Qualit

16、y 24iiPage5.8 The Image Data Block . 255.8.1 Data Structure 256 The Basic Face Image Type 256.1 Inheritance Requirements for the Basic Face Image Type 256.2 Image Data Encoding Requirements for the Basic Face Image Type . 256.3 Image Data Compression Requirements for the Basic Face Image Type . 256.

17、4 Format Requirements for the Basic Image Type . 256.4.1 Facial Header 256.4.2 Facial Information 256.4.3 Image Information 257 The Frontal Face Image Type . 267.1 Inheritance Requirements for the Frontal Face Image Type . 267.2 Scene Requirements for the Frontal Image Type 267.2.1 Purpose . 267.2.2

18、 Pose. 267.2.3 Expression . 267.2.4 Assistance in Positioning the Face 267.2.5 Shoulders. 277.2.6 Backgrounds 277.2.7 Subject and Scene Lighting . 277.2.8 Shadows Over the Face 277.2.9 Shadows in Eye-Sockets . 277.2.10 Hot Spots . 277.2.11 Eye Glasses. 277.2.12 Eye Patches. 287.3 Photographic Requir

19、ements for the Frontal Image Type. 287.3.1 Purpose . 287.3.2 No Over or Under Exposure 287.3.3 Focus and Depth of Field. 287.3.4 Unnatural Color 287.3.5 Color or Grayscale Enhancement 287.3.6 Radial Distortion of the Camera Lens 287.4 Digital Requirements for the Frontal Image Type 297.4.1 Defined F

20、rontal Image Types 297.4.2 Geometry . 297.4.3 Color Profile . 297.4.4 Video Interlacing 307.5 Format Requirements for the Frontal Image Type. 307.5.1 Inheritance Requirements 307.5.2 Image Information 308 The Full Frontal Image Type 308.1 Inheritance Requirements for the Full Frontal Face Image Type

21、 308.2 Scene Requirements for the Full Frontal Face Image Type 30iiiPage8.3 Photographic Requirements for the Full Frontal Face Image Type . 308.3.1 Introduction 308.3.2 Centered Image . 318.3.3 Position of Eyes . 318.3.4 Width of Head 318.3.5 Length of Head. 328.4 Digital Requirements for the Full

22、Frontal Face Image Type. 328.4.1 Resolution 328.5 Format Requirements for the Full Frontal Image Type 328.5.1 Inheritance Requirements 328.5.2 Image Information 329 The Token Face Image Type. 329.1 Inheritance Requirements for Token Face Image Type. 329.2 Digital Requirements for the Token Face Imag

23、e Type. 329.2.1 Introduction 329.2.2 Eye Positions . 339.2.3 Token Image Geometric Format 339.2.4 Minimum Width Token Image 349.2.5 Padding 349.3 Format Requirements for the Token Image Type 359.3.1 Inheritance Requirements 359.3.2 Image Information 35Tables1 Relationships between Facial Image Types

24、 Using the Notion of Inheritance . 22 The Facial Header Block 93 Gender Flags . 104 Eye Color Flags 115 Hair Color Flags . 126 Feature Flags. 137 Expression Flags 148 Center Feature Points 209 The Facial Feature Block . 2010 Facial Image Type Block 2111 Image Data Type Flags 2212 Color Space Flags 2

25、313 Source Type Flags. 2414 The Geometric Characteristics of the Token Image Type 33ivPageFigures1 The Types of Imaging Requirements Specified in this Document . 32 The Facial Image Record Format 73 Definition of Pose Angles. 154 Pose Angles. 165 The Feature Point Set Defined in ISO/IEC 14496-2 186

26、The Eye and Nostril Center Features 197 Image Types and Their Inheritance Map . 228 Geometric Characteristics of the Full FrontalFace Image 319 Token Face Image. 34AnnexesA Best Practices for Frontal Images 36B Best Practices for Full Frontal Images. 42C Best Practices for Token Images. 48D Bibliogr

27、aphy . 53vForeword (This foreword is not part of American National Standard ANSI INCITS 385-2004.)INCITS (The International Committee for Information Technology Standards) is theANSI-recognized Standards Development Organization for information technologywithin the United States of America. Members

28、of INCITS are drawn from Govern-ment, Corporations, Academia and other organizations with a material interest in thework of INCITS and its Technical Committees. INCITS does not restrict membershipand attracts participants in its technical work from 13 different countries, and oper-ates under the rul

29、es of the American National Standards Institute. In the field of Biometrics, INCITS has established the Technical Committee M1. Stan-dards developed by this Technical Committee have reached consensus throughoutthe development process and have been thoroughly reviewed through several PublicReview pro

30、cesses. In addition, this American National Standard has been approvedby the INCITS Executive Board and ANSI Board of Standards Review for Publicationas an ANSI INCITS Standard.This document contains four informative annexes, all of which are not consideredpart of this standard.Requests for interpre

31、tation, suggestions for improvement or addenda, or defect re-ports are welcome. They should be sent to InterNational Committee for InformationTechnology Standards (INCITS), ITI, 1250 Eye Street, NW, Suite 200, Washington,DC 20005.This standard was processed and approved for submittal to ANSI by INCI

32、TS. Com-mittee approval of this standard does not necessarily imply that all committee mem-bers voted for its approval. At the time it approved this standard, INCITS had thefollowing members:Karen Higginbottom, ChairJennifer Garner, SecretaryOrganization Represented Name of RepresentativeApple Compu

33、ter, Inc. David MichaelFarance, Inc Frank FaranceHewlett-Packard Company. Karen HigginbottomScott Jameson (Alt.)Steve Mills (Alt.)EIA Edward Mikoski, Jr.Suan Hoyler (Alt.)IBM Corporation . Ronald F. SillettiInstitute for Certification of Computer Professionals. Kenneth M. ZemrowskiThomas Kurihara (A

34、lt.)IEEE . Judith GormanRichard Holleman (Alt.)Robert Pritchard (Alt.)Intel Corporation . Philip WennblomDave Thewlis (Alt.)Microsoft Corporation . Mike KsarFrank Camara (Alt.)National Institute of Standards Minimize the amount of data needed to be stored for interoperability;Facilitate the use of f

35、ace information with applications that have limited storage;Encourage adoption of biometrics in applications where interoperability is vital;Ensure that enrolled images will meet a quality standard needed for face recog-nition;Improve system throughput by saving the intermediate data instead of the

36、rawdata.AMERICAN NATIONAL STANDARD ANSI INCITS 385-2004 (R2009)American National Standard for Information Technology Face Recognition Format for Data Interchange 1 Document Overview Multiple Face Image Types are proposed to satisfy subsets of function requirements associated with face recognition ap

37、plications briefly discussed in the Introduction: Basic: All face data types shall adhere to the properties of this fundamental type as it specifies the actual data storage format including header and image data format. Photographic (including lighting and pose) or resolution (size or scale) require

38、ments have not been specified for basic images, for flexibility reasons. Frontal: A frontal image is a basic image that adheres to additional photographic requirements appropriate for frontal 2D face recognition and/or human examination. Two types of Frontal images are defined in this document, Full

39、 Frontal and Token Frontal (or simply Token). Full Frontal: This specifies minimal requirements of a grayscale or color face image with sufficient resolution for human examination as well as reliable computer face identification. The image will include the full head with all hair in most cases, as w

40、ell as neck and shoulders. This image is suitable for permanent storage of the face information, and the minimum specifications are to be satisfied for passport, driver license, and “mugshot” images. Token Frontal Image: This is a grayscale or color face image with a specific geometric size and eye

41、positioning based on the width and height of the image. The purpose of this image type is to minimize the storage requirements for computer face recognition tasks such as verification while still offering vendor independence and human verification (versus examination which requires more detail) capa

42、bilities. Other: This image type is reserved for Basic images that do not fall into the Frontal category. ANSI INCITS 385-2004 (R2009) 2 As the face biometric field matures, additional data types may be added to this list through the standards process. For example, one might envision a texture and d

43、epth face data type that would allow for the interchange of information for so-called “3D” face recognition applications. Table 1 Relationships between Facial Image Types Using the Notion of Inheritance Face Data Format Inherits From Normative Clauses Informative Annexes Basic None 1,2,3,4,5,6 None

44、Frontal Basic 7 A Full Frontal Frontal 8 B Token Frontal 9 A3 Other Basic None None NOTE: An example of the relationships between Facial Image Types using the notion of inheritance is as follows: Frontal inherits properties from Basic, which means that all normative clauses that apply to Basic also

45、apply to Frontal. Up to 256x256 images can be stored in a single data record, for the purposes of video capture and processing and modern face recognition enrolment techniques. For each Facial Image Type, four types of requirements are addressed: scene composition, photographic properties, digital p

46、roperties, and the storage of captured information in a data format. The general overview of the requirements is shown in Figure 1. ANSI INCITS 385-2004 (R2009) 3 Requirements Scene Photographic Digitization Data Format Image and Subject Lighting Positioning Camera Attributes Digital Camera Analogue

47、 to Digital Image Scanning Digital Specifications Record Format and Organization Clauses: Clauses: Clauses: Clauses: Basic Face Frontal Face Full Frontal Face None 7.2 8.2 Basic Face Frontal Face Full Frontal Face None 7.3 8.3 Basic Face Frontal Face Full Frontal Face Token Face None 7.4 8.4 9.2 Bas

48、ic Face Frontal Face Full Frontal Face Token Face 5 6.3 6.4 7.5 8.5 9.3 NOTE: The Basic Facial Image Type has no scene, photographic, or digitization requirements. Figure 1 The Types of Imaging Requirements Specified in this Document ANSI INCITS 385-2004 (R2009) 4 1 Scope This biometric data interch

49、ange format specification accommodates: 1) Detailed human examination of facial images 2) Human verification of identity by comparison of persons against facial images 3) Computer automated identification (one-to-many searches) 4) Computer automated verification (one-to-one searches) This standard specifies the proposed record format. The cryptographic protection of the biometrical data structures defined in this document is out of the scope of this standard. 2 Conformance Systems claiming conformance with this standard shall be capable of encoding and decoding