DETECTION AND EXTRACTION OF HUMAN BRAIN TUMOR FROM MRI IMAGE USING IMAGE PROCESSING TECHNIQUE
Abstract
The Magnetic Resonance Image (MRI) is usually associated with impurity and noise that should be removed by the application of image processing technique.
In this study, four stages of image processing technique have been used. At the preprocessing stage, the MRI image was converted to grayscale and the noise was decreased by using a high pass filter. At the enhancement stage, an anisotropic filter was applied to the image to improve the MRI contrast and to remove the noise. At the segmentation stage, the threshold algorithm was used to segment the MRI image. The final stage was applying the morphological operation to the segmented image which allowed the detection and bounding of the tumor in a pure form, as well as forming the contour around the tumor region and thereby separating it from the brain. The area and the execution time of each tumor studied were calculated. When there was no tumor on the MRI image of the brain, the morphological operation provided a sign of its absence. In this study, MATLAB2015a was applied to all stages of image processing technique.
