基于像素滤波和中值滤波的深度图像修复方法

针对Kinect相机采集的深度数据一般有噪声和黑 洞现象,直接应用于场景三维重建系统中效果差的问 题,提出了像素滤波器和中值滤波器相结合的Kinect 深度图像修复方法。首先对一帧图片 上的像素进行搜索, 找到像素值为0的点,以该点为中心,利用其邻域内的像素,定义一个两层的滤波器,根据 像素滤波器的原理 对其进行修复,填充深度图片的黑洞；然后采用中值滤波器,在平滑深度图像的同时保留边 缘信息,去除孤立 噪声点。实验结果表明,本文方法去噪的同时也能对黑洞进行修补,与原始深度图像相比, 空洞明显减少,深度图像质量大大提高。

Depth image inpainting method based on pixel filtering and median filtering

Generally,it has a poor effect if the depth data obtained by Kinect c amera are directly used in 3D-reconstruction system due to the noise and black holes.To sol ve this problem,a depth image restoration method based on pixel filtering and median filtering is proposed.Firstly,search for each pixel in the picture and find the 0value pixels.Secondly,use the pixel filter to repair it and fill the b lack holes in the picture.Finally,bilateral filters are used to remove noise and preserve edge information.The experimental results show that the proposed method can both remove the noise and repair the black holes at the same time.Com pared with the original depth image, the black holes in the picture are obviously reduced,and the image quality is g reatly improved.