This paper presents the Generalized New Entropy (GNE) gait image based authentication system using Information Set concept. A GNE function with free parameters is defined and its properties are presented. A variant of this entropy function called the dynamic entropy function is used in formulating the Dynamic Information Set based Particle Swarm Optimization (DISPSO) technique to learn the parameters. Two types of entropy features called GNE features and GNE based on Histogram of Oriented Gradients (GNE-HOG) features are formulated. After the gait cycle extraction, the farmer features are derived from the probability frames corresponding to the occurrences of 0׳s and 1׳s in every pixel location from all frames (binary silhouette images) contained in a gait cycle whereas the latter features are derived from the HOG descriptors corresponding to the probability frames termed as the possibility frames. The features are validated on three databases (CASIA, OUISIR Treadmill and SOTON small database) using Support Vector Machine (SVM), Euclidean Classifier (EC) and Improved Hanman Classifier (IHC) which is an enhanced version of Hanman Classifier in the literature. The proposed features outperform the existing features using IHC.
|Number of pages||14|
|Publication status||Published - 26-09-2016|
All Science Journal Classification (ASJC) codes
- Computer Science Applications
- Cognitive Neuroscience
- Artificial Intelligence