One of the major challenges in image analysis is image segmentation. In medical applications, skilled operators usually extract the desired regions that may be anatomically separate but statistically indistinguishable. It is subjected to manual errors and biases, which is time consuming, and has poor reproducibility. The problem faced in clustering is the identification of clusters in given data. A widely used method for clustering is based on K-means in which the data is partitioned into K number of clusters. In this method, clusters are predefined which is highly dependent on the initial identification of elements representing the clusters well. Several researchers in clustering has focused on improving the clustering process such that the clusters are not dependent on the initial identification of cluster representation. The proposed method advances an adaptive technique that grows the clusters without the initial selection of elements representing the cluster. It is found to be capable of segmenting the regions of smoothly varying intensity distributions. The technique has been used to achieve an notable accelerated search process.