Classification Human Brain Images and Detection Suspicious Abnormal Area
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2016, Vol 18, Issue 3
Abstract
Abstract: Magnetic Resonance Images (MRI) are widely utilized in the diagnosis of brain images. In this study we have developed a new approach for automatic classification of the normal and abnormal MRI images. Theproposed method consists of four stages namely preprocessing, transformation, texture feature extraction and classification. The first stage is preprocessing image and utilized filter is applied for noise reduction and tomake the image suitable for transformation. In the second stage, is utilized Discrete Multiwavelet Transform (DMWT) to reduce the dimensionality. In the third stage, features extraction based combined between first order statics (FOS) and second order statics (SOS). The fourth stage in the classification stage, a supervised probabilistic neural network (PNN) classifier is utilized to classify the experimental images into normal and abnormal. Finally, proposed algorithm is to segmenting, hexagonal superpixel algorithm. The basic idea in this proposed algorithm is how to detection suspicious abnormal (tumor) area brain to reduce computational time for clinical high diagnostic. It was found this algorithm was very efficient segmentation image, which can be very powerful in the biomedical field of tumor classification.
Authors and Affiliations
Ata'a A. Hasan , Dr. Dhia A. Jumaa
Keywords
Related Articles
- EP ID EP90720
- DOI -
- Views 95
- Downloads 0
Comparative Study between DCT and WaveletTransform Based Image Compression Algorithm
Abstract: Discrete cosine transform emerged as a popular mathematical tool in past decade and widely used inimage compression algorithm due to its high energy compaction capacity. But this energy compaction has som...
Identify the surgically altered face images using granular-PCA approach
Abstract: Plastic surgery provide a way to enhance the facial appearance. The non-linear variations introduced by the plastic surgery has raised a challenge for face recognition algorithms. In this research we ma...
M-Wallet Security using Cued Click Points
Abstract: The main goal of this paper is to provide higher level of security by using graphical password authentication scheme. Text passwords are easily broken by various intruders and may hack all confidential in...
Yours Advance Security Hood (Yash)
[sub][/sub] Hacking of any online account is a practice done by an unauthorized person to get an access in an account of some other person. People says that usually person get hacked because of his own mistake of...
Intelligent Fire Extinguisher System
The multisensor fire-detection algorithm is one of the current important issues in the field of firedetection systems for intelligent buildings. This paper proposes an adaptive fusion algorithm for fire detection,...