The general phenomenon for Image Classification is based on the Feature extraction mechanism. In every domain of image analysis, the classification accuracy is dependent on how better the feature set is generated which helps the machine to learn and predict the unknown sample class label. In this paper, a novel feature extraction mechanism is proposed and named as Counting Label Occurrence Matrix (CLOM). CLOM is based on the counting label of the gray level intensity values of an image and used to extract the textural features of an image for image classification. Four different orientations are used in CLOM for extracting features based on anticipated algorithm. Proposed feature extraction mechanism is dynamic in nature and is used in any domain of image processing and machine learning. CLOM algorithm is compared with some feature extraction algorithms like GLCM (gray level co-occurrence matrix) and Run length Matrix when CLOM is tested in medical domain for classification of brain tumor images. The experimental result shows that the proposed algorithm gives better accuracy than past algorithms when tested with classifiers like KNN and BPNN.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.