A Mathematics Morphology Based Algorithm of Obstacles Clustering

As a large amount of data stored in spatial databases, people may like to find groups of data which share similar features. Thus cluster analysis becomes an important area of research in data mining. In the real world, there exist many physical obstacles such as rivers, lakes and highways, and their presence may affect the result of clustering substantially. However, most of clustering algorithms can not deal with obstacles. In this paper, a new clustering algorithm MMO is proposed for the problem of clustering in the presence of obstacles. The main contributions are: two new mathematics morphological operators are introduced to discover clusters in the presence of obstacles. Our new operators are more accurate than the ordinary operators: open and close. The performance tests show that: MMO is effective in discovering clusters of arbitrary shape in the presence of obstacles; it is very efficient with a complexity of O(N+M) , where N is the number of data points, and M is the number of obstacles; it is not sensitive to noise.