Implementation of Medical Image Segmentation Using K-means Clustering Technique on Fpga

K-Means is an clustering algorithm that is most essential functional to distinctive applications together with color clustering and image segmentation. The dimension of cluster numbers in embedded systems, hardware architecture of hierarchical KMeans (HK-Means) is planned to maintain a maximum cluster number of 1024. A hierarchical memory structure is incorporated to offer a highest bandwidth of 1280 bit/cycle to giving out elements. Features such as video segmentation and color quantization can be executersupport on the planned HK-Means hardware works and associated works. The earlier K-Means architectures cannot make happy the costs of together the computational time and the hardware area. Inembedded systems cost isall the time essential and the effort between the computational time and the hardware area becomes severe particularly as the cluster number enlarge. The large cluster number is a specific design challenge for K-Means hardware architectures. To correct the problem, a new hardware architecture support on hierarchical K-Means (HKMeans) is planned. The presentarchitecture containa hierarchical memory structure to accumulate the cluster centroids for distance calculations and binarytree traversal are used to identify the distant centroid operations in pipeline.