In recent times, Brain Tumors (BT) becomes the deadliest disease, which has increased the mortality rate over the globe. The optimal multilevel thresholding based segmentation and classification model is developed to diagnose BT. Initially, it performs preprocessing in three levels to enhance the quality of the image. Next, multi-level thresholding with artificial bee colony (ABC) algorithm is used for image segmentation. Afterward, the Gray Level Co-occurrence Matrix (GLCM) technique is employed as a feature extractor to extract a useful set of feature vectors. At last, the classification process takes place using a Support Vector Machine (SVM). A set of simulations were carried out using the benchmark Kaggle dataset and the experimental outcome is investigated under different aspects. The results demonstrate the effective diagnostic performance of the presented model with the sensitivity of 97.90%, specificity of 97.91%, and accuracy of 97.56%.
An Optimal Multilevel Thresholding based Segmentation and Classification Model for Brain Tumor Diagnosis
05.11.2020
544376 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Framework for efficient optimal multilevel image thresholding
British Library Online Contents | 2009
|Multilevel thresholding for image segmentation through Bayesian particle swarm optimization
Kraftfahrwesen | 2012
|British Library Online Contents | 2008
|