Selected Quality Metrics for Digital Passport Photographs
This thesis deals with the development of metrics for the automated determination of the quality and grade of acceptance of digital passport photographs without having any reference image available. Different kinds of algorithms were implemented to determine values for image attributes and to detect the face features. No priority for the evaluation of attributes was given in the documents of the international standards. For that reason an international online and on-site survey was developed to explore the opinion of user experts whose work is related to passport photographs. Three different metrics, expressed by the Photograph/ Image and Biometric Attributes Quality Indexes (PAQI, IAQI, BAQI) have been developed to obtain reference values for the quality determination of a passport photograph. Another metric developed is called "Non-Conformance Quality Index" which is based on the representation of the quality information in the minimum unit of information storage: the byte. The nonconformance of a quality attribute is stored in a bit. For a digital passport photograph the representation of the quality attributes is defined by four bytes. Every byte has eight bits and every bit represents an attribute.
What people are saying - Write a review
We haven't found any reviews in the usual places.
Analysis of Conformance Sentences
Categorization of Conformance Requirements Based on an International Survey
Determination of Reference Values for Biometric Quality Attributes
List of Publications
Achto Cualli active contour algorithms analysis Annex Anthropometric Attributes Quality Index background BAQI biometric attributes biometric content blur noise brightness bytes calculated characteristics classiﬁcation color space Colorspace compressed conformance requirements constraints contrast Cualli dataset database deﬁned deﬁnition Design of Experiments determined digital passport photograph Edge detection evaluation Experts face detection face image quality face recognition facial image ﬁeld Figure ﬁle type ﬁnd ﬁrst format Formula head pose angle holder’s horizontally centered face Hue saturation lightness IAQI ICAO identiﬁcation image quality attributes Imax implemented impulsive noise Individual’s face inﬂuence ISO/IEC Length of head method mouth NCQI Non-Conformance Quality Index number of faces obtained perception personal identity documents photograph quality Photograph X1 pixels practical user-experts quality metric red eye Reference Values relevance skin color speciﬁed speciﬁed Not speciﬁed subset survey Table Technical User technical user-experts thesis vertical position visual perception weight Width of head ZHOU WANG