Indian Journal of Science and Technology
Year: 2023, Volume: 16, Issue: 31, Pages: 2398-2408
Original Article
Tanmoy Kanti Halder1, Kanishka Sarkar2*, Ardhendu Mandal3, Bikramadittya Bagchi4
1Assistant Professor, Computer Science, Prasannadeb Women’s College, India
2Assistant Professor, Computer Science, Ananda Chandra College, India
3Associate Professor, Computer Science and Engineering, University of North Bengal, India
4Assistant Teacher, Computer Science, Techno India Group Public School, Hill Cart Road, Darjeeling, India
*Corresponding Author
Email: [email protected]
Received Date:10 April 2023, Accepted Date:25 June 2023, Published Date:14 August 2023
Objectives: To develop a better method for enhancing head MRI or CT images with the objective of suppressing irrelevant information during subsequent processing. Methods: To achieve this, two innovative image enhancement techniques, ’AHE’ and ’Adaptive Gamma with CDF-based Geometric Transformation,’ are combined. The proposed method is evaluated using a total of 76 MRI and 137 CT scan images obtained from the TCIA, OASIS, BCIIHM, and Brain metastases datasets. To tailor these methods for head MRI and CT images, two background masking algorithms have been introduced in this regard. Furthermore, the dark, grey, and white segments of the histogram are identified and geometrically transformed into square areas, followed by a gamma transformation applied to each transformed segment. Finally, a combination of global gamma transformation and AHE is applied to achieve the final enhancement. The proposed method is compared with other state-of-the-art techniques based on evaluation parameters such as ”contrast, ”correlation, and ”entropy. Findings: Both qualitative and quantitative analysis exhibit how the proposed method has better enhancement capability than the CLAHE, AGC, and EGC methods. Our approach yields the highest average contrast, correlation, and entropy values for MRI, measuring 0.522319529, 0.957321634, and 5.585280467, respectively. For CT images, the suggested approach produces the maximum entropy value, 3.472600537, and the average contrast, 0.151943828, which is just less than CLAHE. The average correlation for CT, 0.98799312, is also a hair less than EGC. These results suggest that the proposed method could potentially make the details and structures in the MRI and CT more distinguishable. Novelty : The suggested approach uses a 2D geometric transformation for enhancing contrast of brain CT/MRI images and two robust head masking processes for removing noise in the background. The comparative analysis shows that the proposed method exhibits superior contrast enhancement capability compared to the other three methods being compared.
Keywords: Adaptive power law; CDF based Geometric Transformation; CT Enhancement; Head Masking; MRI Enhancement
© 2023 Halder et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Published By Indian Society for Education and Environment (iSee)
Subscribe now for latest articles and news.