Bio-acoustic signal identification based on sparse representation classifier. (April 2020)

• Record Type: Journal Article
• Authors: Ramli, D A; Wan, Z X; Ibrahim, H
• Is Part Of: Journal of physics. Volume 1529:Number 2(2020)
• Publication Date: 2020-04
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1529/2/022069
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Most animals produce sounds as a way of communication within their species or as noises resulting from feeding or travelling. An automated recognition of bio-acoustic signals becomes vital in the aspect of ecological research and environmental monitoring. With the improvement of technology, scientists today are interested in classifying types and species of animals by their vocalizations without visualizing them with the naked eye. Hence, species identification system based on animal vocalizations becomes an important topic to be researched nowadays. This project aims to develop a frog species voice identification system, recognizing different frog species through analyzing their calls. In this project, Sparse Representation Classifier (SRC) and Kernel Sparse Representation Classifier (KSRC) are employed for the identification task. Performances between SRC and KSRC are compared and discussed in this project. Besides, a graphical user interface (GUI) is also developed to facilitate the user while interacting with the system. Two experiments were done in this project so as to evaluate the effect of numbers of training data and feature dimensions to the classifiers. In short, KSRC (96.6667%) has a higher performance in accuracy compared to SRC (95.6667%). However, KSRC takes a longer computation time compared to SRC. A GUI namely Frog Species Identifier that has been developed by implementing KSRC with 20 training data and 4096 feature dimensions is the final outcome of this project.