摘要
SF6断路器是电力系统中常见的电气设备,针对SF6断路器气体泄漏红外图像中广泛存在的激光散斑噪声,提出一种抑制散斑噪声的新算法。将同态均值滤波与小波变换结合,并应用贝叶斯软阈值,对红外图像进行去噪处理。研究中,将该算法与同态均值滤波、同态均值滤波与传统固定小波阈值结合滤波等算法作了比较,结果表明,在图像标准偏差、信噪比和散斑指数等常用性能指标上所提算法优于其他算法。
SF6 switcher is a common electrical device in power system. Aiming at the laser speckle noise existing widely in the gas leakage infrared image of SF6 switcher, an image de-noising method is proposed in this paper. Homomorphic mean filter, wavelet transform and Bayesian soft threshold are adopted to de-noise the infrared image. In the study, the proposed algorithm is compared with homomorphic mean filter and homomorphic mean filter combined with traditional wavelet threshold denoising algorithm. Experimental results show that in terms of standard deviation, signal to noise ratio （SNR） and speckle index, the proposed algorithm is superior to other existing algorithms.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2011年第12期2669-2673,共5页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金项目(41076060)资助
关键词
气体泄漏
红外图像
散斑噪声
同态均值滤波
小波变换
gas leakage
infrared image
speckle noise
homomorphism mean filter
wavelet transform
