Low and Mid-Level Features for Target Detection in Satellite Images

Provided by: International Journal of Advanced Research in Computer Engineering & Technology
Topic: Mobility
Format: PDF
Reliably detecting objects in broad-area overhead or satellite images has become an increasingly pressing need, as the capabilities for image acquisition are growing rapidly. The problem is particularly difficult in the presence of large intra-class variability. An automatic approach is used to detect and classify targets in high-resolution broad-area satellite images, which relies on detecting statistical signatures of targets, in terms of a set of biologically-inspired low-level visual features. Biologically-inspired feature extraction methods use the "Gestalt" information (continuity, symmetry, closure, repetition) to conduct object detection. Gestalt psychology studies how the human visual system organizes the complex visual input into unitary elements. The goal of the visual system, computer or biological, is to transform a visual input into meaningful semantic information.

Find By Topic