Feature exploration for biometric recognition using millimetre wave body images

Abstract

The use of millimetre wave images has been proposed recently in the biometric field to overcome certain limitations when using images acquired at visible frequencies. Furthermore, the security community has started using millimetre wave screening scanners in order to detect concealed objects. We believe we can exploit the use of these devices by incorporating biometric functionalities. This paper proposes a biometric recognition system based on the information of the silhouette of the human body, which may be seen as a type of soft biometric trait. To this aim, we report experimental results on the BIOGIGA database with four feature extraction approaches (contour coordinates, shape contexts, Fourier descriptors and landmarks) and three classification methods (Euclidean distance, dynamic time warping and support vector machines). The best configuration of 1.33 % EER is achieved when using contour coordinates with dynamic time warping.

Image-based gender estimation from body and face across distances

“…Future work will test the proposed algorithm on less constrained databases that include pose variations. Further, we will explore other features recently employed for body-based people identification such as contour coordinates [15] and will study more sophisticated fusion schemes [14]. by the French Government (National Research Agency, ANR) through the "Investments for the Future" Program reference #ANR-11-LABX-0031-01.…”

Body Shape-Based Biometric Person Recognition from mmW Images

“…We have decided to use the Dynamic Time Warping (DTW) distance, which turned out to be better than other distances and classifiers like SVM in previous works [12]. The goal of DTW is to find an elastic match among samples of a pair of sequences that minimize a given distance measure [19].…”

“…Taking into account the interest of the security community in these frequencies and the promising results obtained in the past few works of recognition through mmW [10,12], one may consider the possibility of using mmW images acquired from the screening scanners simultaneously for detecting hidden objects and performing human recognition. With this approach, the security and recognition control procedures, indispensable for a society threaded constantly, would be enhanced.…”

Exploring Body Texture From mmW Images for Person Recognition

“…However, the experimental protocol carried out in that work was really optimistic, being far away from a realistic verification scenario in which a travel er would enter in a mm W scanner to verify his identity in comparison to a previously enrolled sample. On the other hand, the works by Moreno-Moreno et al [8] and by Gonzalez-Sosa et al [9] proposed and analyzed a biometric person recognition system using shape information, based on the idea that shape information retrieved from mm W images may be more robust to clothing variations than if it were extracted from visible images. In the mentioned works, shape information was extracted from BIOGIGA database [8], which contains synthetically generated mm W images.…”