BRAIN TUMOR SEGMENTATION USING TWO PATH CNN
Ms.P.Vimala Computer Science and Engineering & Dhirajlal Gandhi College of Technology
Ms.S.Megavardhini Computer Science and Engineering & Dhirajlal Gandhi College of Technology
Ms.S.Pooja Computer Science and Engineering & Dhirajlal Gandhi College of Technology
Ms.S.Sriswetha Computer Science and Engineering & Dhirajlal Gandhi College of Technology
Ms.S.Vasumathi Computer Science and Engineering & Dhirajlal Gandhi College of Technology
---------------------------------------------------------------------***--------------------------------------------------------------------
Abstract: The brain is the most important organ in the human body which controls the entire functionality of other organs and it also helps in decision-making. It is also responsible for performing daily voluntary and involuntary activities in the human body. The accurate and exact segmentation of brain tumors is difficult to obtain. In particular, we can use the two-way neuronal convolution networks by using Image Processing. Image Processing is the process of transforming an image into a digital form and performing certain operations to get some useful information from that image itself. Brain tumors can be located everywhere in the brain by their definition, in virtually every form, size, and comparison. It uses neural networks to gain insights from the image data on its own, segments the image into multiple slices, and gives them as input to the neural network. This Paper explores the image with an automated learning system that uses versatile CNN (Convolution Neural Network) and effective CNN. This focuses on the creation of a modern network model, loss feature, data improvement and training and learning to increase the efficiency of brain tumor segmentation. It provides an automated segmentation of brain tumors dependent on deep neural convolution networks. The resulting segmentation method is very quick and 90 percent effective.
Key Words: Brain, Image Processing, Segmentation Brain Tumor, Convolution neural network