Descriptor Generation Techniques Based on Hierarchical-Scale and Multi-Orientation Transforms for Analyzing Form and Layout Characteristics in Aerial Data Classification

Remote sensing is a process used for gathering information about an object or phenomenon without making physical contact with the object and it refers to the activities of recording/observing/ perceiving (sensing) objects or events at far away (remote) places. The remote sensing satellites are equipped with sensors looking down to the earth. They are known as the "eyes in the sky". They constantly observe the earth as they go round in predictable orbits and were used to detect and classify objects in the earth by means of propagated signals (e.g. electromagnetic radiation). Remote sensing data utilizes the electromagnetic radiation that is reflected or emitted from the surface of an object. Remote sensors for example cameras or scanners mounted on satellites or aircraft collect data by energy that is reflected from the earth. They are of two types, 1) active, 2) passive. Passive sensors utilize reflected sunlight as a source of radiation and gathers natural radiation that is emitted or reflected by the object or the surrounding areas. On the contrary, active sensors can generate and detect their own source of energy and are used in the region. In remote sensing the satellite sensors, that are not in direct contact with the object are observed. The electromagnetic radiation act as physical career carrying information from the objects / events to sensors through an intervening medium.