Edge Detection Based On Decision-Level Information Fusion And Its Application In Hybrid Image Filtering

Along with the rapid development of information technology, the image processing technology is developed rapidly in recent years. To understand and have a good command of the development trend would have great significance to the image processing.The edge detection algorithms take an important part in image processing. Traditional edge detection algorithms make detection decision under a single criterion, which may perform inefficiently with the change of noise model. In this theie, edge detection algorithms and fusion algorithms are investigated in detail through studying detection algorithm. The main work and achievement is described as follows:1. The basic concepts and importance of information fusion and edge detection are summarized.2. The concept of fusion entropy is developed to design an information fusion system for edge detection .The weighting coefficients in the decision combination rule are adaptively determined based on the unbiased estimation of fusion entropy and minimum mean square error principles .This new method can remove redundant information, resolve conflicts between multiple detection decisions, so as to retain and enhance the useful information and obtain more reliable edge detection decision. The method has been compared with the traditional single criterion edge detection methods in the application of image filtering. The experimental results can demonstrate the efficiency of the proposed method.3. As a conclusion, the prospect of edge detection was demonstrated .