|
Fuzzy based dynamic histogram equalization for enhancing quality of registered medical image |
|---|---|
| รหัสดีโอไอ | |
| Creator | T. O. Sunitha |
| Title | Fuzzy based dynamic histogram equalization for enhancing quality of registered medical image |
| Contributor | R. Rajalakshmi, S. S. Sujatha |
| Publisher | Rangsit University |
| Publication Year | 2565 |
| Journal Title | Journal of Current Science and Technology |
| Journal Vol. | 12 |
| Journal No. | 2 |
| Page no. | 243-264 |
| Keyword | Contrast enhancement, FEDHE, Medical image, More informative, MSVD, Quality, Registration |
| URL Website | https://jcst.rsu.ac.th/ |
| ISSN | 2630-0656 |
| Abstract | Contrast enhancement is considered as significant aspects in medical analysis because the diagnosis error can be minimised only on utilizing better quality image. In addition to, for performing more accurate and detailed analysis the concept of registration is also included. Many scholars had been focusing on registration approaches to perform their research because of its surplus requirement in medical field. But, the existing registration technique lack in preserving the complementary information and quality of images. Thus, to maintain the features of the original image, the proposed method is designed through coupling contrast enhancement technique along with registration. Initially, in this proposed work, the scanned medical image is taken as input. To improve the quality of image fuzzy employed dynamic histogram equalization is utilized. In this approach initially the histogram of the input image is partitioned based on fuzzy rule. Then, the resultant sub-histogram are equalized based on dynamic range. The final output of this approach is quality enhanced image. Then, registration of this quality enhanced image is attained based on Multi-Scale Singular Value Decomposition (MSVD). The entire proposed architecture is implemented in the MATLAB platform, and the process is compared with the existing approach. Some of the performance metrics, such as average mean brightness error, image contrast, information content and execution time for the proposed method is 0.3345, 242, 5.6 and 36.14 sec. This analysis shows superior performance of the proposed registration induced with contrast enhancement on comparison with existing. Based on this proposed approach high quality image is obtained with less information loss to support medical experts for accurate diagnosis. |
