摘要
对于电力设备的红外图像自动检测系统,图像去噪是非常关键的。针对传统的小波硬、软阈值函数在去噪时存在的不足,在传统小波硬、软阈值函数的基础上对其进行了改进。改进的阈值函数克服了硬阈值函数不连续的缺陷,改善了软阈值函数具有恒定偏差的不足,并引入了两个变量,具有一定的灵活性。同时还使用了一种新的分层阈值选择函数代替统一阈值方法,以改善实际应用的效果。实验结果表明:改进的小波阈值去噪方法在视觉效果、峰值信噪比和均方误差方面都优于传统的硬、软阈值去噪方法;改进的小波阈值去噪方法可以运用到红外图像自动检测系统中,使系统具有更好的去噪效果。
Image denoising is very critical for infrared image automatic detection system in detecting power equip- ment. The traditional wavelet hard-threshold and soft-threshold functions have some deficiencies in image denoising. The improved threshold function overcomes the shortcoming of the hard-threshold with discontinuous function, and the prob- lem of the permanent bias in soft-threshold function is solved based on the hard-threshold and soft-threshold, and it is flexible for two variables are included. At the same time, a new threshold selection method is used instead of the tradi- tional one to improve practical application effect. The simulation results show that the denoising method proposed has better visual effect, higher peak signal-to-noise ratio and lower mean square error than the traditional hard and soft threshold denoising methods. The improved wavelet threshold denoising method can be used in infrared image automatic detection system to make the system get better denoising effect.
作者
施兢业
刘俊
陆龚琪
SHI Jingye LIU Jun LU Gongqi(Shanghai Dianji University, Shanghai 201306, China)
出处
《光学技术》
CAS
CSCD
北大核心
2017年第2期162-165,共4页
Optical Technique
关键词
图像去噪
电力设备
红外图像
小波阈值去噪
阈值选择
image denoising
power equipment
infrared image
wavelet threshold denoising
threshold selection
