Computer aided system for seismic surveying process using UAVs
Ahmed O. G. Ramadan, Sami El Ferik · 2016
In this paper, a methodology for a texture-based classification between sand and rock images will be proposed, and an object recognition algorithm is added for the sand images. Eighteen types of sand and rock textures have been used to extract different type of features which are hypothetically should help in the classification process. The features discrimination ability were tested and ranked using t-test and signal to noise ratio measurements respectively, and based on those, the features with the strongest discrimination power were selected to be used in the classification process to increase the efficiency and reduce the processing time. A back-propagation neural network classifier was trained using the best-selected features until an accuracy of 100% was reached. Different edge detection object recognition methods where used to detect objects in the sand images which was tested with 18 images and the result was an accuracy of 100%.