Information Fusion Techniques for Segmentation of Multispectral Images

Abstract

Image Fusion is the combining process of relevant information from one, two or more images to create a single image which is more complete than any of the input ones. In this paper it is proposed a new segmentation approach for edge detection in multispectral images using image fusion techniques. It was applied for circular objects detection in multispectral images. First, the multispectral images are pre-processed to reduce noise using a speckle filtering method based on homogeneity histogram analysis and then an edge detection method is applied. The results are combined to obtain a more accurate description of the objects in the input images.

Keywords

• Multispectral
• Fusion
• Segmentation
• Edge