A Novel Object Detection Approach Based on the Boundary Shape Information from High Resolution Satellite Imagery

摘要

This paper presents a novel approach of detecting special objects from high resolution satellite imagery. In this approach, a bilateral filtering is used to denoise firstly, and a new morphological approach which combines gray scale morphological processing and binary morphological processing is proposed for ROI extraction and feature enhancement. A detection operator based on the boundary shape information (BSI) is developed to detect enhanced objects. The experiments on images from Google Earth are discussed in the paper. The experimental results show that proposed approach is effective and feasible. Compared with other object detection approaches from high resolution satellite imagery such as PCA or MSNN, the proposed approach has better detection performance.