BRAIN TUMOUR DETECTION OF MRI IMAGES USING MACHINE VISION TECHNIQUES

Abstract

Brain Tumour MRI detection is a challenging task for a doctor. Early detection of a brain tumour disease is a very good for brain tumour patient for the recovery of brain disease. Machine vision techniques are becoming so popular now a days in brain tumour detection. In this research image processing methodology is used for the enhancement of images of brain tumour and clearly detection of MRI images. CVIP tool is also important part of this research the primary purpose of the CVIP tools development environment is to enable students, teachers, researchers, and all users to explore the power of digital image processing. In this research CVIP tool used for image pre-processing, feature extraction, and classifications for result. Through this the results for classification of brain tumour images give better understanding and better results. Different classifiers are used that include Naive Bayes with accuracy 90%, BOVM based SVM with accuracy 97.3% and CNN with accuracy is 98.5%. So, this Accuracies shows that CNN gives a better accuracy with 98.5% than other classifiers.