摘要
针对三维视频图像的空洞填充中的前景背景分割时容易造成前景对象提取不准确而影响修复效果的问题,提出利用梯度融合与聚类相结合的三维视频图像修复算法.首先利用分水岭算法与标记相结合的办法对图像进行分割;然后充分利用深度图像的深度梯度结构信息,并采用K均值聚类对梯度图像进行标记修正,以增强对前景对象的辨别能力.实验结果表明,该算法较好地克服了原有分水岭算法在图像分割过程中易发生过分割现象,完整地提取了前景对象的纹理信息,使修复图像具有更好的视觉效果,峰值信噪比相比于原算法提高了1~3 d B.
For the foreground and background segmentation in three-dimensional video inpainting,the inaccuracy of the foreground extraction is easily to affect the quality of the repaired image.In order to solve this problem,this paper proposes a new three-dimensional video inpainting algorithm based on the combination of gradient fusion and cluster.Firstly,the proposed algorithm combines watershed algorithm with marker for image segmentation,making full use of the structural information of the depth image.Then in order to enhance the ability of distinguishing the foreground objects,K-means clustering is introduced to mark them in the gradient image.The experimental results show that the improved algorithm overcomes the defect of original watershed algorithm in image segmentation,which is likely to occur the over-segmentation phenomenon,and it also can completely extract the texture information of the foreground object,making the repaired image have a better visual effect,and the peak signal-to-noise ratio(PSNR)is increased by 1 to 3 dB,compared to the other algorithm.
作者
来伊丽
唐向宏
楼幸欣
Lai Yili;Tang Xianghong;Lou Xingxin(School of Communication Engineering,Hangzhou Dianzi University,Hangzhou 310018;School of Information Engineering,Hangzhou Dianzi University,Hangzhou 310018)
出处
《计算机辅助设计与图形学学报》
EI
CSCD
北大核心
2018年第3期477-484,共8页
Journal of Computer-Aided Design & Computer Graphics
关键词
分水岭
深度图像
K均值聚类
前景标记
三维视频图像修复
watershed-algorithm
depth image
K-means clustering
foreground marker
three-dimensional video inpainting