Article Preview
TopIntroduction
The increased use of sonar and its difficulties motivated the researcher to produce cost effective and automated process for classification. The recent development of learning algorithm Gorman (Jade, 2013) and al. Networks were trained to classify sonar returns from an undersea metal cylinder and a cylindrically shaped rock. In this paper, we present a new heuristic approach inspired from black holes’ phenomenon based on distance calculation to classify these data.
Active sonar uses a sound transmitter and a receiver. When the two are in the same place, it is monostatic operation. When the transmitter and receiver are separated, it is biostatic operation. When more transmitters (or more receivers) are used, again spatially separated, it is multi-static operation. A beam former is usually employed to concentrate the acoustic power into a beam, which may be swept to cover the required search angles. Figure 1 shows the transmission and receiving process of sonar signals. (Tan, 2004)
Figure 1. Principle of active sonar
Nature is considered as one of the most important, great and immense source of inspiration for solving hard and complex problems in computer science since it exhibits extremely diverse, dynamic, robust, complex and fascinating phenomenon. Over the last few decades, it has stimulated many successful algorithms and computational tools for dealing with complex and optimization problems. Scientist has gone long away inspiring algorithms from some space phenomenon.
This paper presents a new heuristic algorithm for underwater objects identification, it is inspired from the black hole phenomenon. The outlined of this paper is given as follow: section 2 gives a stat of the art to illustrate the relation between machine learning, heuristics and our study before giving an idea about related works in sonar data classification and black holes inspired algorithms. Section 3 presents the black hole phenomenon as Schwarzschild gave. Section 4 details the proposed approach. Finally, section 5 illustrates the used dataset and obtained results compared with other approaches.