A Novel Spatial Clustering with Obstacles and Facilitators Constraint Based on Edge Detection and K-Medoids

Abstract

Clustering spatial data is a well-known problem that has been extensively studied. Grouping similar data in large 2-dimensional spaces to find hidden patterns or meaningful sub-groups has many applications such as satellite imagery, geographic information systems, medical image analysis, marketing, computer visions, etc. Spatial clustering has been an active research area in Spatial Data Mining (SDM). Many methods on spatial clustering have been proposed in the literature, but few of them have taken into account constraints that may be present in the data clustering. In this paper, we discuss the problem of spatial clustering with obstacles constraints and propose a novel spatial clustering using edge detection method and K-Mediods, which objective is to cluster the spatial data (images) with the constraints and also comparing the result with the various constraints based clustering algorithms in terms of number of clusters and its execution time.The Edge detection based K-Mediods algorithms can not only given attention to higher speed and stronger global optimum search, but also get down to the obstacles and facilitator constraints and practicalities of spatial clustering. Taking into account these constraints during the clustering process is costly and the modeling of the constraints is paramount for good performance. The results on real datasets shown that the Edge detection based spatial clustering with the constraints are performs better than the existing constraint based clustering.

Medical Images Segmentation Based on Unsupervised Algorithms: A Review

“…A helpful tool is a silhouette to define k. As compared to k-mean, it is more potent for outliers and noise. Medoids can be defined as a mass object in which the average variation is negligible with all objects in the block, i.e., In the stated data set, this is the focal point [16].…”

“…Unsupervised algorithms that cluster objects into groups increase the similarity between objects in a cluster and minimize the similarity between objects in various categories. This fascinating problem has been interesting for several years because of its many implementations [16]. Examples of unsupervised learning algorithms systems include (PAM, K-Mean).…”

A Comparative approach to cluster maternal data using k-means and k-medoid

“…The method is implemented in real-time GPS applications for detecting the road lanes from satellite data. (Pattabiraman et al, 2009) Have proposed a new spatial clustering method for edge detection. This paper discusses the obstacles related to spatial clustering in detail.…”