A Hybrid Feature Extraction for Satellite Image Segmentation Using Statistical Global and Local Feature

Satellite image segmentation is a principal task in many applications of remote sensing such as natural disaster monitoring and residential area detection. These applications collect a number of features of an image and according to different features of an object will detect the object from the image. This type of image (satellite image) is rich and various in content, the most of methods retrieve the textural features from various methods but they do not produce an exact descriptor features from the image. So there is a requirement of an effective and efficient method for features extraction from the image, some approaches are based on various features derived directly from the content of the image. This paper presents a new approach for satellite image segmentation which automatically segments image using a supervised learning algorithm into urban and non-urban area. The entire image is divided into blocks where fixed size sub-image blocks are adopted as sub-units. We have proposed a fusion of statistical features including global features based on the common moment of RGB image and local features computed by using the probability distribution of the phase congruency computed on each block. The results are provided and demonstrate the good detection of urban area with high accuracy and very fast speed.