Indian Journal of Radio & Space Physics

Open Access Open Access  Restricted Access Subscription Access

Renyi entropy based Bi-histogram equalization for contrast enhancement of MRI brain images


Affiliations
1 Department of Electronics and Comunication Engineering, National Institute of Technology Puducherry, Karaikal 609 609, India
 

The quality of the MRI brain images is dependent on the sensor. It is essential to have a pre-processing technique to meet the finest quality at the sensor’s cost. A pre-processing algorithm has been proposed in this paper to enhance the low contrast MRI brain images. The input image’s histogram has been divided into two sub histograms using its median value to uphold the input image’s mean brightness. After calculating the Renyi entropy from the sub histogram, histogram clipping has been done to regulate the enhancement rate. The clipping limit has been selected automatically from the minimum value of the mean, median of the distribution function, and itself. Additionally, the proposed algorithm has incorporated the Discrete Cosine Transform (DCT) to improve the enhancement. Experimental results have shown that the proposed algorithm enhances the input image and maintains the mean brightness.
User
Notifications
Font Size

Abstract Views: 105

PDF Views: 70




  • Renyi entropy based Bi-histogram equalization for contrast enhancement of MRI brain images

Abstract Views: 105  |  PDF Views: 70

Authors

Vijayalakshmi D
Department of Electronics and Comunication Engineering, National Institute of Technology Puducherry, Karaikal 609 609, India
Poonguzhali Elangovan
Department of Electronics and Comunication Engineering, National Institute of Technology Puducherry, Karaikal 609 609, India
Malaya Kumar Nath
Department of Electronics and Comunication Engineering, National Institute of Technology Puducherry, Karaikal 609 609, India

Abstract


The quality of the MRI brain images is dependent on the sensor. It is essential to have a pre-processing technique to meet the finest quality at the sensor’s cost. A pre-processing algorithm has been proposed in this paper to enhance the low contrast MRI brain images. The input image’s histogram has been divided into two sub histograms using its median value to uphold the input image’s mean brightness. After calculating the Renyi entropy from the sub histogram, histogram clipping has been done to regulate the enhancement rate. The clipping limit has been selected automatically from the minimum value of the mean, median of the distribution function, and itself. Additionally, the proposed algorithm has incorporated the Discrete Cosine Transform (DCT) to improve the enhancement. Experimental results have shown that the proposed algorithm enhances the input image and maintains the mean brightness.