Trung Nguyen Tu
Faculty of Computer Science and Engineering, Thuyloi University, 175 Tay Son, Dong Da, Hanoi, 010000, Vietnam.
Curr Med Imaging. 2023 Jul 20. doi: 10.2174/1573405620666230720103039.
Image enhancement is a very significant topic in image processing that improves the quality of the images. The methods of image enhancement are classified into 3 categories. They include histogram method, fuzzy logic method and optimal method. Studies on image enhancement are often based on rules: if it's bright then it's brighter, if it's dark then it's darker and using the global approach. So, it's hard to enhance objects in all dark and light areas, as in the medical images.
Input data is downloaded from the link: http://www.med.harvard.edu/AANLIB.
This paper introduces a new algorithm for enhancing medical images that is called the medical image enhancement based on cluster enhancement (MIECE). Firstly, the input image is clustered by the algorithm of fuzzy clustering. Then, the upper bound, and lower bound are calculated according to cluster. Next, the sub-algorithm is implemented for clustering enhancement using an enhancement operator. For each pixel, the gray levels for each channel (R, G, B) are transformed with this sub-algorithm to generate new corresponding gray levels. Because after clustering, each pixel belongs to each cluster with the corresponding membership value. Therefore, the output gray level value will be aggregated from the enhanced gray levels by the sub-algorithm with the weight of the corresponding cluster membership value.
This paper experiences the method MIECE with input data downloaded from the link: http://www.med.harvard.edu/AANLIB. The experimental results are compared with some recent methods that include: SGHIE (2017), Ying (2017) and KinD++ (2021).
This paper introduces the new algorithm which is based on cluster enhancement (MIECE) to enhance the medical image contrast. The experimental results show that the output images of the proposed algorithm are better than some other recent methods for enhancing dark objects.
图像增强是图像处理中一个非常重要的主题，它可以提高图像质量。图像增强方法分为三类，包括直方图方法、模糊逻辑方法和优化方法。图像增强的研究通常基于规则：如果图像亮则使其更亮，如果图像暗则使其更暗，并采用全局方法。因此，在医学图像中，很难增强所有暗区和亮区的物体。
从链接http://www.med.harvard.edu/AANLIB下载输入数据。
本文介绍了一种新的医学图像增强算法，即基于聚类增强的医学图像增强算法（MIECE）。首先，使用模糊聚类算法对输入图像进行聚类。然后，根据聚类计算上界和下界。接下来，使用增强算子实现聚类增强子算法。对于每个像素，使用该子算法对每个通道（R、G、B）的灰度值进行变换，以生成新的对应灰度值。因为聚类后，每个像素以相应的隶属度值属于每个聚类。因此，输出灰度值将由子算法根据相应聚类隶属度值的权重从增强后的灰度值中聚合得到。
本文介绍了一种基于聚类增强的新算法（MIECE）来增强医学图像对比度。实验结果表明，该算法的输出图像在增强暗物体方面优于其他一些近期方法。
