![Open Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltextgreen.png)
![Restricted Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltextred.png)
![Open Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltextgreen.png)
![Open Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltext_open_medium.gif)
![Restricted Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltextred.png)
![Restricted Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltext_restricted_medium.gif)
A Novel Approach for MRI Brain Image Segmentation using Local Independent Projection Model
Subscribe/Renew Journal
Brain tumor segmentation is an important process for tumor identification and assists to planning for further treatment. Although several brain tumor segmentation methods are existing, still efficient brain tumor segmentation is challenging in medical field. To achieve the high detection accuracy of tumor part with lower error rate, we have lots of enhancing techniques for the tumor segmentation methods. In this paper, we propose a Novel approach for brain Image Segmentation using Local Independent Projection Model for MRI images. The main objective of this paper is to develop a system for brain segmentation based on local independent projection classification. The proposed system has 2 stages such as training and testing and it has 4 steps such as pre-processing, feature extraction, segmentation and post processing. Preprocessing is done before starting process. In feature extraction the related features from the input data to be retrieved. This project proposed the patch based method used for feature extraction. Then apply the local independent projection classification. The segmentation of brain tumor can be assumed as a multiclass classification problem. Resolving this problem by One-Versus-All (OvA) strategy In this strategy, a classifier is trained for each class to differentiate a class from all other classes. In this classification first construct the dictionary based on original samples in training set. Then present the sparse representation using locally linear representation. Dictionary construction is performed by using manually labeled original samples in a training set. In order to achieve classification scores, Softmax regression model is used. By using learned as well as without learned softmax regression model, classification accuracy was tested. Finally calculate the classification score computation. After post processing to get the final results.
Keywords
![](https://i-scholar.in/public/site/images/abstractview.png)
Abstract Views: 182
![](https://i-scholar.in/public/site/images/pdfview.png)
PDF Views: 1