Classification of Remotely Sensed Data by Texture Features with the Nature-Inspire Optimization Algorithm
S. Soja Rani · International Journal for Research in Applied Science and Engineering Technology · 2019
At regular intervals, the satellite remote sensing technologies collect images/data and volume of information gathered is substantial and it is developing exponentially as the innovation is developing at a quick speed. Remote sensing technique has been extensively utilized for recognition of land use and land cover structures. In this paper, we have used four classes of images which are enhanced by Prewitt filter and features are extracted by texture approaches, which increases the pattern visibility. The proposed methodology here includes an efficient feature extraction and further classification of the satellite images on the basis of these features using Particle Swarm Optimization, hybrid (PSO+ACO). For the training Convolution Neural Network (CNN) is used. The various performance metrics have been determined and contrasted according to the recovered satellite images from the dataset.