Francesco Nicolo

Long Range Cross-Spectral Face Recognition: Matching SWIR Against Visible Light Images

Short wave infrared (SWIR) is an emerging imaging modality in surveillance applications. It is able to capture clear long range images of a subject in harsh atmospheric conditions and at night time. However, matching SWIR images against a gallery of color images is a very challenging task. The photometric properties of images in these two spectral bands are highly distinct. This work presents a novel cross-spectral face recognition scheme that encodes images filtered with a bank of Gabor filters followed by three local operators: Simplified Weber Local Descriptor, Local Binary Pattern, and Generalized Local Binary Pattern. Both magnitude and phase of filtered images are encoded. Matching encoded face images is performed by using a symmetric I-divergence. We quantify the verification and identification performance of the cross-spectral matcher on two multispectral face datasets. In the first dataset (PRE-TINDERS), both SWIR and visible gallery images are captured at a close distance (about 2 meters). In the second dataset (TINDERS), the probe SWIR images are collected at longer ranges (50 and 106 meters). The results on PRE-TINDERS dataset form a baseline for matching long range data. We also demonstrate the capability of the proposed approach by comparing its performance with the performance of Faceit G8, a commercial face recognition engine distributed by L1. The results show that the designed method outperforms Faceit G8 in terms of verification and identification rates on both datasets.

A method for robust multispectral face recognition

Matching Short Wave InfraRed (SWIR) face images against a face gallery of color images is a very challenging task. The photometric properties of images in these two spectral bands are highly distinct. This work presents a new cross-spectral face recognition method that encodes both magnitude and phase of responses of a classic bank of Gabor filters applied to multi-spectral face images. Three local operators: Simplified Weber Local Descriptor, Local Binary Pattern, and Generalized Local Binary Pattern are involved. The comparison of encoded face images is performed using the symmetric Kullbuck-Leibler divergence. We show that the proposed method provides high recognition rates at different spectra (visible, Near InfraRed and SWIR). In terms of recognition rates it outperforms Faceit R - G8, a commercial software distributed by L1.

Cross spectral iris matching based on predictive image mapping

An adaptive method to predict NIR channel image from color iris images is introduced. Both visual inspection of the predicted image and the verification performance indicate that the adaptive mapping linking NIR image and color image is a potential solution to the problem of matching NIR images vs. color images in practice. When matched against NIR enrolled image the predicted NIR image achieves significantly high performance compared to the case when the same NIR image is matched against R channel alone.

On Empirical Recognition Capacity of Biometric Systems Under Global PCA and ICA Encoding

Performance of biometric-based recognition systems depends on various factors: database quality, image preprocessing, encoding techniques, etc. Given a biometric database and a selected encoding method, the recognition capability of a system is limited by the relationship between the number of classes that the recognition system can encode and the length of encoded data describing the template at a specific level of distortion. In this paper, we evaluate empirical recognition capacity of biometric systems under the constraint of two global encoding techniques: principal component analysis (PCA) and independent component analysis (ICA). The developed methodology is applied to predict the capacity of different recognition channels formed during the acquisition of different iris and face databases. The proposed approach relies on data modeling and involves classical detection and information theories. The major contribution is in providing a guideline on how to evaluate capabilities of large-scale biometric recognition systems that are based on PCA and ICA encoding. Recognition capacity can also be promoted as a global quality measure of biometric databases.

Adaptive biometric authentication using nonlinear mappings on quality measures and verification scores

Three methods to improve the performance of biometric matchers based on vectors of quality measures associated with biometric samples are described. The first two methods select samples and matching scores based on predicted values of Quality of Sample (QS) index (defined here as d-prime) and Confidence in matching Scores (CS), respectively. The third method treats quality measures as weak but useful features for discrimination between genuine and imposter matching scores. The unifying theme for the three methods consists of a nonlinear mapping between quality measures and the predicted values of QS, CS, and combined quality measures and matching scores, respectively. The proposed methodology is generic and is suitable for any biometric modality. The experimental results reported show significant performance improvements for all the three methods when applied to iris biometrics.

Encoding, matching and score normalization for cross spectral face recognition: Matching SWIR versus visible data

We propose a methodology for cross matching color face images and Short Wave Infrared (SWIR) face images reliably and accurately. We first adopt a recently designed image encoding and matching technique which is capable to encode face images in both visible and SWIR spectral bands. Encoding is performed in two steps. Images are initially filtered with a bank of Gabor filters. Then three local operators: Simplified Weber Local Descriptor and Local Binary Pattern applied to magnitude of filtered images and Generalized Local Binary Pattern applied to the phase are involved to create histogram-like feature templates. The distance between two encoded face images is measured by symmetric I-divergence. The encoding and matching methods are demonstrated on long range SWIR data matched against close range visible images. A considerable performance improvement is observed compared to the results by FaceIt G8. To further enhance performance we propose an adaptive score normalization approach. We demonstrate that significant performance improvement is achieved with a small training set. Matching scores obtained by the proposed normalized method and by FaceIt G8 are fused to result in further performance improvement.

A method for selecting and ranking quality metrics for optimization of biometric recognition systems

In the field of biometrics evaluation of quality of biometric samples has a number of important applications. The main applications include (1) to reject poor quality images during acquisition, (2) to use as enhancement metric, and (3) to apply as a weighting factor in fusion schemes. Since a biometric-based recognition system relies on measures of performance such as matching scores and recognition probability of error, it becomes intuitive that the metrics evaluating biometric sample quality have to be linked to the recognition performance of the system. The goal of this work is to design a method for evaluating and ranking various quality metrics applied to biometric images or signals based on their ability to predict recognition performance of a biometric recognition system. The proposed method involves: (1) Preprocessing algorithm operating on pairs of quality scores and generating relative scores, (2) Adaptive multivariate mapping relating quality scores and measures of recognition performance and (3) Ranking algorithm that selects the best combinations of quality measures. The performance of the method is demonstrated on face and iris biometric data.

Iris Quality Metrics for Adaptive Authentication

Iris sample quality has a number of important applications. It can be used at a variety of processing levels in iris recognition systems, for example, at the acquisition stage, at image enhancement stage, or at matching and fusion stage. Metrics designed to evaluate iris sample quality are used as figures of merit to quantify degradations in iris images due to environmental conditions, unconstrained presentation of individuals or due to postprocessing that can reduce iris information in the data. This chapter presents a short summary of quality factors traditionally used in iris recognition systems. It further introduces new metrics that can be used to evaluate iris image quality. The performance of the individual quality measures is analyzed, and their adaptive inclusion into iris recognition systems is demonstrated. Three methods to improve the performance of biometric matchers based on vectors of quality measures are described. For all the three methods, the reported experimental results show significant performance improvement when applied to iris biometrics. This confirms that the newly proposed quality measures are informative in the sense that their involvement results in improved iris recognition performance.

Near real-time face detection and recognition using a wirelesscamera network

We present a portable wireless multi-camera network based system that quickly recognizes face of human subjects. The system uses low-power embedded cameras to acquire video frames of subjects in an uncontrolled environment and opportunistically extracts frontal face images in real time. The extracted images may have heavy motion blur, small resolution and large pose variability. A quality based selection process is first employed to discard some of the images that are not suitable for recognition. Then, the face images are geometrically normalized according to a pool of four standard resolutions, by using coordinates of detected eyes. The images are transmitted to a fusion center which has a multi-resolution templates gallery set. An optimized double-stage recognition algorithm based on Gabor filters and simplified Weber local descriptor is implemented to extract features from normalized probe face images. At the fusion center the comparison between gallery images and probe images acquired by a wireless network of seven embedded cameras is performed. A score fusion strategy is adopted to produce a single matching score. The performance of the proposed algorithm is compared to the commercial face recognition engine Faceit G8 by L1 and other well known methods based on local descriptors. The experiments show that the overall system is able to provide similar or better recognition performance of the commercial engine with a shorter computational time, especially with low resolution face images. In conclusion, the designed system is able to detect and recognize individuals in near real time.

Empirical capacity of a biometric channel under the constraint of global PCA and ICA encoding

The ability of practical biometric systems to recognize a large number of subjects is constrained by a variety of factors that include a choice of a source encoding technique, quality of images, complexity and variability of underlying patterns and of collected data. Given a source encoding technique, the remaining factors can be attributed to distortions due to a biometric recognition channel. In this work, we define empirical mutual information and recognition rate and evaluate empirical recognition capacity of biometric systems under the constraint of two global encoding techniques: principal component analysis (PCA) and independent component analysis (ICA). The empirical capacity of biometric systems is numerically evaluated as a point of intersection of the empirical mutual information rate curve plotted as a function of the recognition rate and the diagonal line bisecting the first quadrant. The developed methodology is applied to find the empirical capacity of different recognition channels formed during acquisition of different iris and face databases.