Object Recognition In Underwater Imaging Using Machine Learning Techniques

Abstract

For many years, mines in the ocean are becoming major worry and threat to newlinehuman lives and vessel safety. These mines are generally placed in the ocean newlinefor security reasons to protect from enemies which can destroy submarines newlineand ship which comes in contact with the mines. It s very difficult to identify newlineand detect the objects in underwater using sonar imagery because of its newlinecomplications. This is due to factors which involve variations in operational newlineand environment conditions, spatially variable chaos, variation in target newlineshapes, structure and orientation. Considering all these conditions, a method newlinehad been proposed which can detect and classify whether the object is a mine newlineor an object which resembles a mine under water. Images are obtained from newlinesonar camera scanner, which is placed in underwater communication newlinenetwork in a moving vehicle with a sensor. In our application, using image newlineprocessing and machine learning technologies, we have studied the newlinebehaviour and differentiate the features of a mines and rocks. In most cases newlinemines are considered to be metal objects. We have designed this application newlinewith algorithm which can give a real time capability to detect the objects and newlinedistinguish them as seabed objects and imaginary artifacts which are induced newlineby vehicles. UCI dataset is considered in this system which holds all possible newlineattack data with percentage of possibility. This data is then utilized in the newlinepre-processing by applying one hot encoding. Statistical methods are used to newlinev newline newlineobtain Z-scores, mean, median and mode for determining the significant newlinefeatures and training on 80% of dataset is validated. The remaining 20% newlinedataset is tested and validated using Decision tree, K-NN and Gradient newlineboosting algorithm. The efficiency of these algorithms is being analysed and newlinediscussed and it is found that Gradient boosting algorithm is best suitable newlinealgorithm to be utilized for development of mine detection. Different newlineparameters were interacted based on the testing and checked for false newlinene