Multi-Biometric and Multi-Modal Authentication

Dear Colleagues,

Extensive studies have been performed on several biological traits, regarding their capacity to be used for unimodal biometric recognition. When analysing a biometric trait considering the four critical factors: universality, uniqueness, collectability, and permanence, a couple of general conclusions can be withdrawn: (i) there is no “gold-standard” biometric trait, i.e., the choice of the best biometric trait will always be conditioned by the means at our disposal and the specific application of the recognition process; and (ii) some biometric traits seem to present advantages that counterbalance other trait’s disadvantages. Marked advantages might be found by exploring the synergistic effect of multiple statistically independent biometric traits. Using biometric evidence obtained from multiple sources of information will result in an improved capability of tackling some of the more relevant known problems of unimodal systems such as dealing with noisy data; intra-class variations; inter-class similarities; non-universality; and presentation attacks.

The three biometric traits considered by the standards developed within the International Civil Aviation Organization (ICAO) are face, fingerprints, and iris. Face is the primary biometric trait chosen for most states, nevertheless any state can provide additional data input to the identity verification processes by including multiple biometrics in their travel documents, i.e., a combination of face and/or fingerprint and/or iris.

Like in other fields of research, the existence of suitable databases is crucial for the development and evaluation of methodologies. Multimodal biometric databases were not always available for several reasons. To overcome this difficulty, some works use chimeric or virtual-subject databases with several drawbacks such as the fact that these databases contain users that do not exist in the real world which the system will never encounter. This practice also disregards one of the goals of fusion systems which is to describe the relationship among different modalities.

In general, user authentication can be performed based on one or a combination of the following items: (i) something the user knows (e.g., password, personal identification number (PIN), secret answer, pattern); (ii) something the user has (e.g., smart card, ID card, security token, software token, smartphone); and (iii) something the user is or does (e.g., fingerprint, face, iris, gait). As a premise, it is worth considering that passwords can be forgotten or snatched by malicious people; physical objects, such as badges and ID documents, can be lost or stolen. Biometrics can hardly be stolen, and the process of falsification is much more complicated. If we consider all possible combinations of the three factors of authentication, higher security can be achieved by combining biometric traits with another authentication factor.

This Special Issue of Journal of Imaging aims to feature reports of recent advances in multi-biometric and multi-modal authentication that are specifically designed to overcome unimodal biometric systems’ limitations. We particularly encourage the submission of work that adopts innovative biometric traits (as opposed to the ICAO traits that are already widely used) used in combination with ICAO and non-ICAO biometrics or authentication factors. Other topics of interest include new fusion techniques, as well as new multi-biometric databases (including artificially generated ones) that would make it possible to use machine learning techniques based on deep learning, which are known to require large amounts of data. Finally, we recommend the use of international ISO/IEC standards for the presentation of results.

Dr. Chiara Galdi
Dr. Daria La Rocca
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Journal of Imaging is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• multi-modal authentication
• multi-biometric authentication
• multi-instance biometrics
• fusion strategies
• innovative biometrics
• multi-modal biometric database