The Impact of Preprocessing on Deep Representations for Iris Recognition on Unconstrained Environments

Abstract

The use of iris as a biometric trait is widely used because of its high level of distinction and uniqueness. Nowadays, one of the major research challenges relies on the recognition of iris images obtained in visible spectrum under unconstrained environments. In this scenario, the acquired iris are affected by capture distance, rotation, blur, motion blur, low contrast and specular reflection, creating noises that disturb the iris recognition systems. Besides delineating the iris region, usually preprocessing techniques such as normalization and segmentation of noisy iris images are employed to minimize these problems. But these techniques inevitably run into some errors. In this context, we propose the use of deep representations, more specifically, architectures based on VGG and ResNet-50 networks, for dealing with the images using (and not) iris segmentation and normalization. We use transfer learning from the face domain and also propose a specific data augmentation technique for iris images. Our results show that the approach using non-normalized and only circle-delimited iris images reaches a new state of the art in the official protocol of the NICE.II competition, a subset of the UBIRIS database, one of the most challenging databases on unconstrained environments, reporting an average Equal Error Rate (EER) of 13.98% which represents an absolute reduction of about 5%.

Deep representations for cross‐spectral ocular biometrics

“…Recently, machine learning techniques based on deep learning have been achieving great popularity due to the results reported in the literature, which advance the state-of-the-art in various problems such as speech recognition [8][9][10], natural language processing [11,12], digit and character recognition [13,14] and face recognition [15,16]. In the field of ocular biometrics, using deep learning representation has been advocated both for the periocular [17,18] and iris [6,[19][20][21][22][23][24][25] regions, with interesting and promising results being reported.…”

“…The strategy described in this paper is composed of some methodologies extracted from the literature. For both the iris and ocular traits, we use as input the bounding box delimited regions used in the state-of-the-art methods [17,25]. Then, the features from these traits were extracted using a similar approach proposed by Zanlorensi et al [25].…”

Simultaneous Iris and Periocular Region Detection Using Coarse Annotations

“…An overview of the important features of all databases used in this work can be seen in Table II. These databases were chosen because they are widely used in the biometric recognition literature [23], [27]- [29], which we plan to investigate in future works.…”

“…Typically, in biometric systems that use iris and/or periocular region images as input, the first step in which efforts should be applied is the detection of the Region of Interest (ROI) [21], as a poor detection would probably impair the effectiveness of the subsequent steps of the system [12], [22]. Recently, Zanlorensi et al [23] showed that impressive iris recognition rates can be achieved when using deep representations having as system input the bounding boxes of the iris region, without the iris segmentation preprocessing. Also using deep representations and having as input a squared region (i.e., a bounding box), Luz et al [24] achieved state-of-the-art results for periocular recognition.…”

“…However, despite the good results presented by that segmentation approach, the detection task is much less expensive in terms of both computational cost and data annotation. Regarding the 11 databases employed in our experiments, they were chosen because they are widely used in the biometric recognition literature [23], [27]- [29], which we plan to investigate in future works. It should be noted that, in many works in the literature, no more than three databases were used in the experiments [30]- [33].…”