EFFICIENT NEURON SEGMENTATION AND MORPHOLOGICAL ANALYSIS USING SVM-KNN CLASSIFIER

Abstract

Listen

Translate article

A novel method of neuron segmentation in image volumes attained by microscopy is a proposed model. Like agglomerative or correlation clustering, existing methods relay solely on boundary signs and have problems where such a piece of evidence is lacking (e.g., incomplete staining) or ambiguous (e.g., co-located cell and mitochondria membranes). This paper investigates these complications through sparse region appearance cues distinguish between pre-and postsynaptic neuron segments in neural tissue. In this proposed paper, for pre-processing stage, an adaptive histogram equalization technique and wiener filters employed. Efficient segmentation performed by Morphological processing and Otsu’s threshold. Using support vector machine (SVM) classifier to classify tumours (Neuroblastoma) and the results compare with the prevailing-nearest neighbour (KNN) model. SVM classifier achieves higher accuracy, sensitivity, specificity, precision, Recall, gmean than the existing KNNmodel.