Biometric identification via eye movement scanpaths in reading

Abstract

This paper presents an objective evaluation of various eye movement-based biometric features and their ability to accurately and precisely distinguish unique individuals. Eye movements are uniquely counterfeit resistant due to the complex neurological interactions and the extraocular muscle properties involved in their generation. Considered biometric candidates cover a number of basic eye move ments and their aggregated scan path characteristics, in cluding: fixation count, average fixation duration, average saccade amplitudes, average saccade velocities, average saccade peak velocities, the velocity waveform, scanpath length, scanpath area, regions of interest, scan path inflec tions, the amplitude-duration relationship, the main se quence relationship, and the pairwise distance between fIXations. As well, an information fusion method for com bining these metrics into a single identification algorithm is presented. With limited testing this method was able to identifY subjects with an equal error rate of 27%. These results indicate that scan path-based biometric identifica tion holds promise as a behavioral biometric technique.

The Impact of Temporal Proximity between Samples on Eye Movement Biometric Identification

“…Such a recording is easy to analyze, because it requires that fixations and saccades happen in specific moments. However, there are several interesting experiments with different scenarios, including faces observation [25] or text reading [11]. There is also an attempt to perform identification without any information about a stimulus [18].…”

A Model of Individual Differences in Gaze Control During Reading

“…Models are estimated on the eye movement records of individuals on the training sentences (Equation 8 (full model), a model variant in which the variable d t+1 and corresponding distribution is removed (saccade type + amplitude), and a simple model that only fits a multinomial distribution to saccade types (saccade type only). Additionally, we compare against the feature-based reader identification approach by Holland & Komogortsev (2012). Six of the 14 features used by Holland & Komogortsev depend on saccade velocities and vertical fixation positions.…”

Biometric Authentication Using Mouseand Eye Movement Data

“…To achieve this they trained a system by extracting features for eye velocity, eye movement direction, and distance traveled. [13] was able to achieve an FAR and FRR of 27%. To reach this accuracy they used data gathered from users reading text.…”