期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2017
卷号：5
期号：6
页码：12538
DOI：10.15680/IJIRCCE.2017.0506202
出版社：S&S Publications
摘要：This paper is concerned with the detection and classification of object in underwater video. Identifyingand arranging objects in frames is a vital submerged application with pertinence to maritime transportation and barrier.By taking into consideration the limitations of side scan sonar images, often introduce large infraclassvariability’s dueto imaging modalities that reduces the discriminative power of any classification algorithm and limiting thepossibilities of improving classification accuracy. So, to make the proposed method the robust one we are here usingthe k means clustering for the segmentation andDNN (Deep Neural Network) classifier is used for the classificationpurpose.According to the features extracted by method HOG ,we can classify the object based on RGB plane into twoclasses.In this proposed system, we are classifying the objects as ‘Dynamic’ and ‘Static’.
