Automated detection in microscopic images using segmentation

In this paper, we present a segmentation clusteringbased approach for automated object detection. This paper deals with the segmentation and classification of blood cells for the purpose of detecting leukemia (abnormal blood cells). After the image acquisition and the preprocessing step, we proceeded to the application of the k-means method. In order to show the interest of the proposed approach, we present the different cancerous regions identified with their characteristics for biomedical diagnostic aid. The proposed method is tested on image dataset and achieves 98% segmentation accuracy. These results show that our approach offers encouraging performance and best automatic leukemia detection. The proposed system is successfully implemented in Matlab, experimental results demonstrate that our approach offers encouraging performance and better quality automatic leukemia detection.