The Study Of Boundary Detecting Algorithm Based On Joint Entropy

With the increasing development of information technology and database technology, the data category and quantity in the database demonstrates a sharp growth. Therefore, how to obtain effectively the valuable information from the mass data is of great significance. Meanwhile, in order to satisfy the requirement, data mining technology arises at the historic moment. Suitable data mining methods make it possible for biologists to discover the massive genetic information and geographers to discover atmospheric pressure pattern in term of polar region and the sea which has a remarkable influence on land climate. As one of the important data mining, clustering technology has already been done thorough research thus presenting many kinds of cluster algorithms. However, the research on cluster boundary's algorithms is still in its infancy. Cluster boundary, a kind of mode, is widely applied in practice. For instance, it represents the outline of the object in the image detection and healthy people having some potential diseases in the clinical medicine. Therefore, the research on cluster boundary is of great importance.To better the existing algorithm, this paper brings forth an Efficient boundary points Detecting alGorithm based on joint Entropy (EDGE) and Boundary points detecting clusterring Algorithm based Gradient Binarization (BAGB).The EDGE algorithm withdraws the cluster boundary points in combination of grid technology and joint entropy technology. Grid technique is used to fast search the scope of grids where clusters boundary exists, thus shrinking search scope and improving algorithm efficiency. Joint entropy is applied to detect boundary points of clusters in these grids and therefore increases algorithm precise. The experimental results show that EDGE can precisely detect boundary of clusters in terms of different shapes, sizes and density of the data sets and effectively eliminating noises. Meanwhile, the time complexity of algorithm is linear function of input the data set, so the superiority of execution time is more obvious in the large-scale data set.The BAGB algorithm, combining grid technology and gradient operator, withdraws the cluster boundary points. In this algorithm the grid technology is used to enhance the speed of the data processing, and prewitt gradient operator is applied to calculate gradient in 3×3 grid region from eight directions, the maximum being the central grid gradient. The gradient is used to judge whether the grid is the boundary grid or not, and a point in the boundary grid is a boundary point.Putting the method of image boundary processing into the practice of processing cluster boundary is a fresh idea for the research on cluster boundary. The experimental results indicate that this algorithm can effectively detect boundary of clusters in datasets with noises/outliers and improve operating sufficiency.The innovations of this paper are to put forward, for one thing, the idea of detecting cluster boundary based on the combination of networking and the union entropy technology and then provide EDGE algorithm, for another, the thought of detecting cluster boundary through combining grid and gradient operator and furnish BAGB algorithm.