Search for Articles:

MECS Press Menu

Souli S. Sameh

Work place: Signal, Image and pattern recognition research unit, ENIT/ Dept. of Genie Electrique BP 37, 1002, Le Belvédère, Tunisia

E-mail: soulisameh@yahoo.fr

Website:

Research Interests: Image Processing, Image Manipulation, Image Compression, Computer systems and computational processes

Biography

Souli S. Sameh Birth in Tunis, on 26/12/1980. Assistant at Higher Institute of Arts and Multimedia Manouba (ISAMM), Tunis. Have a thesis degree in image and signal processing on 14 March 2014 at national school of Engineers of Tunis (ENIT), Tunisia. She has a master in Electronics: Automatic and signal processing, National School of Engineers of Tunis (ENIT), Tunis, Tunisia in October 2008. She has a master's degree in Computer Science, Faculty of Sciences of Tunis (FST), Computer Science Department, Tunisia in June 2005. She is Assistant for six years at Higher Institute of Arts and Multimedia Manouba (ISAMM), she has a journal paper named: ―Multiclass support vector machines for environmental sounds classification in visual domain based on log-Gabor filters‖ at Int J Speech Technol, 2013. She has a book chapter named: ―Environmental Sounds Classification Based on Visual Features‖ at CIARP, 2011. Dr. Souli is a review in Science Publishing Group journal.

Author Articles
On the Use of Time–Frequency Reassignment and SVM-Based Classifier for Audio Surveillance Applications

By Souli S. Sameh Zied LACHIRI

DOI: https://doi.org/10.5815/ijigsp.2014.12.03, Pub. Date: 8 Nov. 2014

In this paper, we propose a robust environmental sound spectrogram classification approach. Its purpose is surveillance and security applications based on the reassignment method and log-Gabor filters. 
Besides, the reassignment method is applied to the spectrogram to improve the readability of the time-frequency representation, and to assure a better localization of the signal components. Our approach includes three methods. In the first two methods, the reassigned spectrograms are passed through appropriate log-Gabor filter banks and the outputs are averaged and underwent an optimal feature selection procedure based on a mutual information criterion. The third method uses the same steps but applied only to three patches extracted from each reassigned spectrogram. The proposed approach is tested on a large database consists of 1000 sounds belonging to ten classes. The recognition is based on Multiclass Support Vector Machines.

[...] Read more.
Other Articles

Subscribe to receive issue release notifications and newsletters from MECS Press journals