基于改进MSR的小波变换图像增强算法
Wavelet Transform Image Enhancement Algorithm Based on Improved MSR
DOI: 10.12677/CSA.2021.114118, PDF,  被引量   
作者: 陈宏辉, 彭向前:湖南科技大学机电工程学院,湖南 湘潭;胡小平:湖南科技大学先进矿山装备教育部工程研究中心,湖南 湘潭
关键词: 改进MSR小波变换同态滤波图像增强Improved MSR Discrete Wavelet Transform Homomorphic Filtering Image Enhancement
摘要: 针对采集的触摸屏图片存在图片整体偏暗、对比度低,背景噪声较多,难以分割图像中缺陷的问题,提出一种基于改进MSR (多尺度Retinex滤波)的小波变换图像增强算法。该算法首先对图像进行小波分解,得到图像的低频分量使用以中值滤波模板代替高斯滤波模板的改进MSR处理以增强低频系数,对高频分量进行同态滤波以限制噪声,最后通过小波逆变换重建图像。本文将改进的算法应用于采集的触摸屏表面缺陷图片,通过VS2017软件进行仿真实验,并与同态滤波算法、MSR算法进行对比。实验结果表明,使用改进MSR算法提高了算法的运行速度,该算法优于传统的MSR算法,对图像的增强、噪声的抑制有良好的效果,改善了图像的整体视觉效果且大大缩短了算法的运行时间。
Abstract: As for the problems existed in the Touch screen picture such as relatively dark image as a whole, low contrast, background noise interference and difficulties in segmenting the defects in image, a wavelet transform image enhancement algorithm based on improved Multi-Scale Retinex (MSR) is proposed. The algorithm first performs Wavelet Transform on image, does improved MSR using median filter template instead of Gaussian filter template on the low frequency component of image to enhance the low-frequency coefficient, finishes homomorphic Filter on the high frequency component to limit the noise, and finally reconstructs the image by inverse wavelet transform. This paper applies improved algorithm to image enhancement and carries out simulation experiments by using multiple images. The algorithm proposed in this paper is realized by VS2017 software programming and is compared with homomorphic filter algorithm and the MSR algorithm. The experimental results show that the use of the improved MSR algorithm improves the running speed of the algorithm. This algorithm is better than the HOMOMORPHIC filtering algorithm and MSR algorithm and has a good effect on image enhancement and noise suppression, improves the overall visual effect of the image and greatly shortens the running time of the algorithm.
文章引用:陈宏辉, 胡小平, 彭向前. 基于改进MSR的小波变换图像增强算法[J]. 计算机科学与应用, 2021, 11(4): 1149-1156. https://doi.org/10.12677/CSA.2021.114118

参考文献

[1] 杨茂祥. 低照度环境下彩色图像增强算法研究[D]: [硕士学位论文]. 南京: 南京邮电大学, 2019.
[2] Subramani, B. and Veluchamy, M. (2020) Quadrant Dynamic Clipped Histogram Equalization with Gamma Correction for Color Image Enhancement. Color Research & Application, 45, 644-655. [Google Scholar] [CrossRef
[3] Zhuang, P. and Din, X. (2020) Correction to: Underwater Image Enhancement Using an Edge-Preserving Filtering Retinex Algo-rithm. Multimedia Tools and Applications. 79, Article No. 17279. [Google Scholar] [CrossRef
[4] 王伟江, 彭业萍, 曹广忠, 郭小勤. 面向机柜表面缺陷检测的不均匀光照和低亮度图像增强方法[J]. 仪器仪表学报, 2019, 40(8): 131-139.
[5] 陈茹霞, 强振平, 邵小锋, 何丽波. 基于L0范数的Retinex图像增强算法[J]. 计算机工程与科学, 2020, 42(7): 1244-1252.
[6] 王彦, 谢晓方, 肖楚琬, 刘明春, 翟胜路, 张晓瑜. 基于改进MSR滤波算法的X光图像增强[J]. 计算机工程, 2012, 38(8): 186-188.
[7] 冯红波, 李萍, 王博. 基于自适应权重Retinex和小波变换的彩色图像增强算法[J]. 无线电工程, 2020, 50(1): 28-33.
[8] Xia, K.-J., Wang, J.-Q. and Cai, J. (2019) A Novel Medical Image Enhancement Algorithm Based on Improvement Correction Strategy in Wavelet Transform Domain. Cluster Computing, 22, 10969-10977. [Google Scholar] [CrossRef
[9] 杨静. 基于小波变换的低对比度图像增强方法[J]. 计算机时代, 2011(1): 10-12.
[10] Wang, W., Li, B., Zhang, J., Xian, S. and Wang, J. (2008) A Fast Multi-Scale Retinex Algo-rithm for Color Image Enhancement. Proceedings of the 2008 International Conference on Wavelet Analysis and Pattern Recognition, Hong Kong, 30-31 August 2008, 30-31. [Google Scholar] [CrossRef
[11] 于天河, 孟雪, 潘婷, 兰朝凤. 小波变换和自适应变换相结合的图像增强方法[J]. 哈尔滨理工大学学报, 2018, 23(6): 100-104.
[12] 贾翔宇. 基于小波变换的夜间低照度图像降噪与增强算法[J]. 信息技术与信息化, 2019(2): 107-109.
[13] 周峡, 徐善顶. 一种改进小波阈值函数的图像去噪方法研究[J]. 南京工程学院学报(自然科学版), 2019, 17(4): 44-49.
[14] 张学典, 杨帆, 常敏. 基于图像信息熵统计直方图的图像增强算法[J]. 包装工程, 2020, 41(13): 251-260.
[15] 李庆忠, 王丰凯. 基于直方图自适应拉伸的水下图像增强算法[J]. 计算机应用研究, 2020, 37(S1): 408-411.