Karim Faten Khalid, Jalab Hamid A, Ibrahim Rabha W, Al-Shamasneh Ala'a R
Department of Computer Science, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, P.O Box 84428, Riyadh 11671, Saudi Arabia.
Department of Computer System & Technology, Faculty of Computer Science, and Information Technology, Universiti Malaya, 50603 Kuala Lumpur, Malaysia.
J King Saud Univ Sci. 2022 Oct;34(7):102254. doi: 10.1016/j.jksus.2022.102254. Epub 2022 Jul 28.
The medical image enhancement is major class in the image processing which aims for improving the medical diagnosis results. The improving of the quality of the captured medical images is considered as a challenging task in medical image. In this study, a trace operator in fractional calculus linked with the derivative of fractional Rényi entropy is proposed to enhance the low contrast COVID-19 images. The pixel probability values of the input image are obtained first in the proposed image enhancement model. Then the covariance matrix between the input image and the probability of a pixel intensity of the input image to be calculated. Finally, the image enhancement is performed by using the convolution of covariance matrix result with the input image. The proposed enhanced image algorithm is tested against three medical image datasets with different qualities. The experimental results show that the proposed medical image enhancement algorithm achieves the good image quality assessments using both the BRISQUE, and PIQE quality measures. Moreover, the experimental results indicated that the final enhancement of medical images using the proposed algorithm has outperformed other methods. Overall, the proposed algorithm has significantly improved the image which can be useful for medical diagnosis process.
医学图像增强是图像处理中的一个主要类别，其目的是改善医学诊断结果。提高所采集医学图像的质量被认为是医学图像领域一项具有挑战性的任务。在本研究中，提出了一种与分数阶Rényi熵导数相关联的分数阶微积分中的迹算子，用于增强低对比度的新冠肺炎图像。在所提出的图像增强模型中，首先获取输入图像的像素概率值。然后计算输入图像与输入图像像素强度概率之间的协方差矩阵。最后，通过协方差矩阵结果与输入图像的卷积来执行图像增强。所提出的增强图像算法针对三个不同质量的医学图像数据集进行了测试。实验结果表明，所提出的医学图像增强算法使用BRISQUE和PIQE质量度量都实现了良好的图像质量评估。此外，实验结果表明，使用所提出的算法对医学图像进行最终增强的效果优于其他方法。总体而言，所提出的算法显著改善了图像，这对医学诊断过程可能是有用的。
