Hybrid Feature Extraction Approach for Robust Brain Tumor Classification: HOG, GLCM, and Artificial Neural Network

The classification of brain tumors is an essential undertaking that helps physicians make precise decisions and treat patients appropriately based on their classes. To make the right decision, it is therefore necessary to improve the classification accuracy. The aim of this study is to develop a hybrid feature extraction method using effective features for classifying accurate brain tumors. Firstly, area labeling and skull stripping with intensity thresholding are used to segment the tumor cell region. Next, the canny algorithm is used to identify the segmented tumor cell zone. The characteristics of this identified tumor cell area are then used as the input of the ANN classification network to do the classification. Also, histogram of oriented gradients (HOG) and gray-level co-occurrence matrix (GLCM) are used for feature extraction. The method that is suggested works better, yielding a 98% accuracy rate. Additionally, it performs better than several other methods that are currently in use that combine hybrid or single feature extraction techniques with various classification algorithms.