Analysis of Gait Flow Image and Gait Gaussian Image Using Extension Neural Network for Gait Recognition

Analysis of Gait Flow Image and Gait Gaussian Image Using Extension Neural Network for Gait Recognition

Parul Arora, Smriti Srivastava, Shivank Singhal
DOI: 10.4018/978-1-7998-0414-7.ch025
(Individual Chapters)
No Current Special Offers


This paper proposes a new technique to recognize human gait by combining model free feature extraction approaches and a classifier. Gait flow image (GFI) and gait Gaussian image (GGI) are the two feature extraction techniques used in combination with ENN. GFI is a gait period based technique, uses optical flow features. So it directly focuses on dynamic part of human gait. GGI is another gait period based technique, computed by applying Gaussian membership function on human silhouettes. Next, ENN has been used as a classifier which combines the extension theory and neural networks. All the study has been done on CASIA-A and OU-ISIR treadmill B databases. The results derived using ENN are compared with SVM (support vector machines) and NN (Nearest neighbor) classifiers. ENN proved to give good accuracy and less iteration as compared to other traditional methods.
Chapter Preview

1. Introduction

Human identification has become a challenging area from the pattern recognition point of view. Over the past few years various biometrics including palm print [Wang X et. al, 2013], iris [Rai H et. al, 2014], face [Luo L et. al, 2015], finger knuckle [Kumar A et. al, 2013] have been worked on for human identification. But the limitation of these modalities is that they use some specific devices to obtain features and their intrusive nature. Unlike the modalities mentioned above, gait a behavioral modality, is a right choice to identify human based on their non-intrusive characteristics. Earlier studies [Cutting J 1977; Murray M et. al, 1964] also show that gait is a unique biometric modality. To obtain gait patterns of a subject, a single camera is enough [Yu et. al, 2014]. Gait recognition can easily identify people at a distance by their walk manner. In today’s scenario, terrorism is the main threat for human security [Inbarani et. al, 2015; Sarkar M et. al, 2015]. Therefore Human gait recognition has become an attractive field for researchers and industrialists from the surveillance point of view [Okumara M et. al, 2010; Sundaresan A et. al, 2003].

Two different approaches have been adopted to recognize gait: model-based and model-free. Model based techniques [Cunado D et. al, 1999; Tafazzoli et. al, 2010; Wang L et. al, 2004; Yam C et. al, 2004] work on structural model of the human body. It has the advantage of being non sensitive to outliers and background noise [Lee L et. al, 2002; Yoo et. al, 2011]. But due to complexity present in model based methods, model free methods are more reliable and easy to explore. Model free methods do not require any structure model; it works on information stored in binary silhouettes [Kale et. al 2004, Sarkar et. al 2005, Xu et. al 2012].

Complete Chapter List

Search this Book: