Using Structural Features of Targets on Classification and Removal of Gap Region in SONAR Images

Seas and theirs risks for humans caused that warning equipments like SONAR attended by many of researchers about seabed. In this paper, a new method is introduced for sonar image processing and removal of gap region in side-scan sonar images. In the proposed method, to improve removal efficiency of gap region and target detection in sonar images, phase congurency algorithm and structural characteristics of shape are used. In this study, Haqqani port, Bandar Abbas, Iran seabed images are used to show the efficiency of phase congruency method. The proposed method is compared with classic gradient method, and the results show that the proposed method has more efficiency than classic methods. These method improved parameters like: sensitivity (4.4 %), Detected Evaluation Ratio (4 %), Detected False Evaluation Ratio (75.6 %) Algorithm running time (5.8 Sec).