An Overview of Bio-Inspired and Deep Learning Model for Extraction of Land Use Pattern

To study land use pattern with satellite images, there are many procedures for image processing. Like feature extraction from many images, image enhancement by removing noise, classification of images, and ultimately image interpretation. Image processing generally has to deal with many complex problems like time-dependent, noise removal, and making the image data at the lowest computational costs with high accuracy with low energy consumption. All these forms very challenging problems and here comes machine learning, deep learning and optimization. Optimization is related to each and every field. It means minimizing the cost and energy inputs and to maximize the efficiency and performance of the system. It also implies proper utilization of resources. In the domain of image processing and optimization recently new methods are emerging like bio-inspired modelling and deep learning techniques. Bio-inspired models are computer algorithms that are based on the biological action of nature like fireflies, swarms, ant colony and bee colony which are used for image feature extraction and classification. In feature extraction, a deep learning process is used in which features are automatically extracted by the intelligence of machines. with all these techniques it gives use a better efficiency than traditional feature extraction.