计算机应用与软件2017，Vol.34Issue(8)：225-230,6.DOI:10.3969/j.issn.1000-386x.2017.08.040
基于快速行进算法的深度图像增强算法
DEPTH IMAGE ENHANCEMENT ALGORITHM BASED ON FAST MARCHING METHOD
孟恬 1全红艳1
作者信息
- 1. 华东师范大学计算机科学与软件工程学院 上海 200062
- 折叠
摘要
Abstract
In order to solve the problem of existing holes in depth images captured by Kinect, we proposed an enhancement algorithm based on the Fast Marching Method.First, we used the color image as input to calculate the Multiple-Scale Structural Similarity of each pixel with their neighbors.Then we used MSSIM to construct the diffusion function, through the improved FMM algorithm, to estimate the depth information to fill depth holes.Finally, to highlight the details of edge features, we used an improved guide filtering algorithm to enhance the edge details.Experimental results show that the algorithm can accurately fill the holes of the depth image and can effectively keep the edge details of the object.With this algorithm, we can solve the problem and can promote the research and application of the 3D reconstruction.关键词
深度图/空洞/FMM/MSSIM/扩散函数/自适应/引导滤波Key words
Depth map/Hole filling/FMM/MSSIM/Spread function/Adaptive/Guided bilateral filtering分类
信息技术与安全科学引用本文复制引用
孟恬,全红艳..基于快速行进算法的深度图像增强算法[J].计算机应用与软件,2017,34(8):225-230,6.
