A robust technique of brain MRI classification using color features and k-nearest neighbor algorithm
The analysis of MRI images is a manual process carried by experts which need to be automated to accurately classify the normal and abnormal images. We have proposed a reduced, three stage model having per-processing feature extraction and classification steps. in pre-processing the noise has been r...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Science & Engineering Research Support soCiety
2016
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/52576/1/Abdul-salam-oct-2016-paper-A-Robust-Technique-of-Brain-MRI-Classification-using-Color-Features-and-K-Nearest-Neighbors-Algorithm.pdf http://irep.iium.edu.my/52576/ http://www.sersc.org/journals/IJSIP/vol9_no10/2.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The analysis of MRI images is a manual process carried by experts which need to be automated to accurately classify the normal and abnormal images. We have proposed a reduced, three stage model having per-processing feature extraction and classification steps. in pre-processing the noise has been removed from gray scale images using a median filter, and than gray sclae images have been converted to color (RGB) images. In feature extraction, red, green and blue channels from each channel of RGB has been extracted because they are so much informative and easier to process. The first three color moments mean, variance, and skewness are calculated for eaqch red, green and blue channel images. The feature extraction in the feature extraction stage are classified into normal and abnormal with K-Nearest Neighbor (K, NN). This method is applied to 100 images ( 70 normal , and 30 abnormal). The proposed method gives 98.00% training and 95% test accuracy with databases of normal images and 100% training and 90% test accuracy with abnormal images. The average computation time for each image was 0.06s. |
|---|