图像直方图分析中的信息粒度。

Information granules in image histogram analysis.

机构信息

Faculty of Biomedical Engineering, Silesian University of Technology, ul. Roosevelta 40, 41-800 Zabrze, Poland.

出版信息

Comput Med Imaging Graph. 2018 Apr;65:129-141. doi: 10.1016/j.compmedimag.2017.05.003. Epub 2017 May 10.

DOI:10.1016/j.compmedimag.2017.05.003

Abstract

A concept of granular computing employed in intensity-based image enhancement is discussed. First, a weighted granular computing idea is introduced. Then, the implementation of this term in the image processing area is presented. Finally, multidimensional granular histogram analysis is introduced. The proposed approach is dedicated to digital images, especially to medical images acquired by Computed Tomography (CT). As the histogram equalization approach, this method is based on image histogram analysis. Yet, unlike the histogram equalization technique, it works on a selected range of the pixel intensity and is controlled by two parameters. Performance is tested on anonymous clinical CT series.

摘要

讨论了一种基于强度的图像增强中使用的粒度计算概念。首先，介绍了加权粒度计算的思想。然后，介绍了该术语在图像处理领域的实现。最后，介绍了多维粒度直方图分析。所提出的方法专用于数字图像，特别是计算机断层扫描 (CT) 采集的医学图像。与直方图均衡方法一样，该方法基于图像直方图分析。然而，与直方图均衡技术不同，它在像素强度的选定范围内工作，并由两个参数控制。在匿名临床 CT 系列上进行了性能测试。