Airport Detection on Remote Sensing Images Using Deep Convolutional Neural Network

This study analyzed usage of deep convolutional neural networks (CNNs) in given that a clarification aimed at the answer of detecting airport in various remote sensing images (RSIs). Last few years, deep convolutional neural networks are mostly used in many uses in the area of Image processing, computer vision as well as remote sensing. So research found detecting the airport is one of the pattern recognition difficulty, in that firstly different features be located are extracted, also then one of the classifier like CNN or SIFT or SVM is used to detect airports. As per the research in various field CNNs gives the improved classification accurateness than various methods exists. The technique suggested in this research firstly detects different areas on RSIs using Hough transform, after that usage of these candidate regions for the training of CNN architecture. The CNN model designed here has five convolutions along with three fully connected layers. Normalization plus dropout layers are present in directive to construct effective design. Here data augmentation approach be situated used in the direction of decrease over fitting as well as enlarges the amount of dataset.