A new thresholding technique based on fuzzy set as an application to leukocyte nucleus segmentation

摘要: A new fuzzy method is introduced to segment leukocytes in blood smear images.Interval-valued intuitionistic set generated by minimizing ultrafuzziness.Similarity between ideally thresholded and the segmented images are computed.Best threshold obtained maximizing similarity.Experimentally proven that proposed better than other methods. Background objectivesThe main aim of this paper using interval-valued sets (IVIFSs). Generally, uncertainties occur terms vagueness through brightness levels image. Processing such uncertain can be efficiently handled sets, particularly IVIFSs. MethodsLogarithmic membership function utilized for computing values corresponding intensities pixel. Non-membership IVIFS constructed Yager generating function. By varying parameters, 256 IVIFSs generated. An selected from having ultrafuzziness along with threshold. Threshold determined finding an maximum similarity ideal results method. ResultsQuantitatively, evaluated precision-recall, receiver operator characteristic curves, Jaccard coefficient measure structural index time taken segmenting nucleus, their compared existing Performance measures reveal seems comparable ConclusionsSegmentation helps analyst differentiating various types determination leukocyte count, counting essential out diseases related reduction or surplus quantity these cells.