Clustering Techniques for Image Segmentation
Buch in deiner Nähe kaufen
Beschreibung
This book presents the workings of major clustering techniques along with their advantages and shortcomings. After introducing the topic, the authors illustrate their modified version that avoids those shortcomings. The book then introduces four modified clustering techniques, namely the Optimized K-Means (OKM), Enhanced Moving K-Means-1(EMKM-1), Enhanced Moving K-Means-2(EMKM-2), and Outlier Rejection Fuzzy C-Means (ORFCM). The authors show how the OKM technique can differentiate the empty and zero variance cluster, and the data assignment procedure of the K-mean clustering technique is redesigned. They then show how the EMKM-1 and EMKM-2 techniques reform the data-transferring concept of the Adaptive Moving K-Means (AMKM) to avoid the centroid trapping problem. And that the ORFCM technique uses the adaptable membership function to moderate the outlier effects on the Fuzzy C-meaning clustering technique. This book also covers the working steps and codings of quantitative analysis methods. The results highlight that the modified clustering techniques generate more homogenous regions in an image with better shape and sharp edge preservation.
- Showcases major clustering techniques, detailing their advantages and shortcomings;
- Includes several methods for evaluating the performance of segmentation techniques;
- Presents several applications including medical diagnosis systems, satellite imaging systems, and biometric systems.
This book presents the workings of major clustering techniques along with their advantages and shortcomings. After introducing the topic, the authors illustrate their modified version that avoids those shortcomings. The book then introduces four modified clustering techniques, namely the Optimized K-Means (OKM), Enhanced Moving K-Means-1(EMKM-1), Enhanced Moving K-Means-2(EMKM-2), and Outlier Rejection Fuzzy C-Means (ORFCM). The authors show how the OKM technique can differentiate the empty and zero variance cluster, and the data assignment procedure of the K-mean clustering technique is redesigned. They then show how the EMKM-1 and EMKM-2 techniques reform the data-transferring concept of the Adaptive Moving K-Means (AMKM) to avoid the centroid trapping problem. And that the ORFCM technique uses the adaptable membership function to moderate the outlier effects on the Fuzzy C-meaning clustering technique. This book also covers the working steps and codings of quantitative analysismethods. The results highlight that the modified clustering techniques generate more homogenous regions in an image with better shape and sharp edge preservation.
- Showcases major clustering techniques, detailing their advantages and shortcomings;
- Includes several methods for evaluating the performance of segmentation techniques;
- Presents several applications including medical diagnosis systems, satellite imaging systems, and biometric systems.
Showcases major clustering techniques, detailing their advantages and shortcomings Includes several methods for evaluating the performance of segmentation techniques Presents several applications including medical diagnosis systems, satellite imaging systems, and biometric systems
Autor*in
Fasahat Ullah Siddiqui
Themen in »Clustering Techniques for Image Segmentation«
Image Segmentation and Clustering Hard and Soft Clustering Techniques Enhanced Clustering Techniques Mathematical Model of clustering techniques Mathematical Model of evaluation methods