基于改进的MSR和γ-CLAHE的骶髂关节CT图像增强算法  

Sacroiliac Joint CT Image Enhancement Algorithm Based on Improved MSR and γ-CLAHE

作　　者：闫建红[1] 李佳欣 YAN Jianhong;LI Jiaxin(College of Computer Science and Technology,Taiyuan Normal University,Jinzhong 030619,China)

机构地区：[1]太原师范学院计算机科学与技术学院,山西晋中030619

出　　处：《现代信息科技》2025年第6期126-129,134,共5页Modern Information Technology

摘　　要：针对骶髂关节疾病患者的个体差异等因素导致CT图像对比度不佳、图像边缘不清晰等问题,设计一种结合改进多尺度Retinex(MSR)和对比度受限的自适应直方图均衡化(CLAHE)的方法,用于增强骶髂关节CT图像。首先,对骶髂关节CT图像进行双边滤波以去除噪声,接着使用带Gamma校正的CLAHE算法对图像进行处理,提升图像的对比度。采用引导滤波与高斯滤波的联合加权作为中心环绕函数的Retinex算法对图像进行处理,以增强图像边缘信息。最终,加权融合两种处理后的图像,得到最终效果图。实验结果表明,该算法在图像信息熵、对比度、峰值信噪比方面显著优于传统算法。To address the issues of poor contrast and unclear image edges in CT images caused by factors such as individual differences among patients with sacroiliac joint diseases,an improved method combining Multi-Scale Retinex(MSR)and Contrast-Limited Adaptive Histogram Equalization(CLAHE)is designed to enhance sacroiliac joint CT images.Firstly,bilateral filtering is used to denoise the sacroiliac joint CT images.Then,the images are processed using the CLAHE algorithm with Gamma correction to enhance image contrast.And it uses the Retinex algorithm which takes the combined weighted of guided filtering and Gaussian filtering as the center-surround function to process the images,so as to enhance edge information.Finally,the two types of processed images are weighted and fused to obtain the final effect images.Experimental results show that this algorithm significantly improves image entropy,contrast,and peak signal-to-noise ratio compared to traditional algorithms.

关 键 词：骶髂关节CT图像 RETINEX CLAHE 图像增强 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]