基于3维轴距与非局部均值的无人机巡检图像去噪算法

A denoising algorithm in aerial transmission line image based on three-dimensional axis distance and non-local means

无人机搭载摄像设备获得输电线路的数字图像,利用数字图像处理技术可以提高故障识别率,而图像在采集和传输过程中易受到噪声干扰,图像质量受损不利于后续输电线故障识别,因而有必要对图像进行去噪处理。以无人机巡检高压输电线图像为对象,针对航拍图像中存在椒盐噪声问题,研究了椒盐噪声的3维直方图分布特点,并且在非局部均值滤波算法的基础上改进相似性权重,提出一种结合3维轴距的非局部均值去噪方法,解决了原非局部均值算法主要针对高斯噪声以及耗时过长的问题。首先,根据椒盐噪声的特点,对含噪图像进行3维轴距预滤波处理,计算图像中每个像素点的3维轴距,依据轴距检测噪声点,并对噪声点进行中值滤波。其次,对噪声点和非噪声点设置不同的权重,噪声点不计入相似性度量中,使得像素块之间相似度判断更加准确,利用改进的权重计算函数作非局部均值去噪。实验结果表明,所提算法能够有效去除巡检图像中的椒盐噪声,尤其是背景复杂图像中有显著的去噪效果,图像细节和边缘信息保持较好。同时所提算法与原非局部均值相比,耗时低,实时性得以提高。

The image of power transmission line can be obtained by the cameras carried with UAV,and the fault identification can be improved by using digital image processing technology.However,images are easily disturbed by noises in the process of generation and transmission,which degrades the quality of images and has unfavorable influence on fault identification,so it is necessary to carry on denoising processing.In order to restore the aerial image corrupted by salt and pepper noise,the 3D histogram distribution was studied.Based on the nonlocal means filter,the similarity weight was improved.An algorithm based on three-dimensional axis distance and non-local means was proposed in this paper for the inspection of high-voltage transmission line with Unmanned Aerial Vehicle,which turns out to be a solution to problem that the NLM was mainly aimed at gaussian noise and high time consuming.Firstly the image was pre-denoising by three-dimensional axis distance.It detected noise pixels by three-dimensional axis distance,the noise points were removed by median filter.Then the different weights were set for noise points and non-noise points,the weights of noise points were zero which made the similarity measurement between pixels blocks more accurate.And finally the image was re-denoised by the NLM with improved weight function.The experimental results demonstrate that the proposed algorithm has good performance in removing salt and pepper noise of Aerial Transmission Line Image,thus enhancing the image edge well.It provided guarantee for the following power line inspection.Besides,it has lower time consumption and improve the real-time performance.