摘要： 鉴于压缩域视频运动分割方法在分割速度上的优越性，提出一种基于H.264的压缩域视频运动对象分割方法，对初始的运动矢量场进行去噪、中值滤波、校正和累积处理，得到更可靠的运动矢量场，用改进的模糊C-均值聚类算法分割出视频序列中的运动对象。实验结果表明，该方法可以快速准确地提取出视频序列中的运动对象。
关键词:
H.264标准,
压缩域,
对象分割,
中值滤波,
运动矢量,
模糊C-均值聚类
Abstract: According to the advantage of segmentation speed in the compressed domain, this paper proposes a method of video object segmentation based on H.264. It eliminates the noise, median filter, correction and accumulation to processing initial motion vectors and uses fuzzy C-means clustering algorithm to process motion vectors and segments the moving objects. Experimental results show that the method can get the moving object fast and accuracy.
Key words:
H.264 standard,
compressed domain,
object segmentation,
median filtering,
motion vector,
fuzzy C-means clustering
中图分类号:
轩腾蛟, 康志伟. 基于模糊聚类的H.264压缩域视频对象分割[J]. 计算机工程, 2012, 38(3): 267-269.
HAN Teng-Jiao, KANG Zhi-Wei. Video Object Segmentation in H.264 Compressed Domain Based on Fuzzy Clustering[J]. Computer Engineering, 2012, 38(3): 267-269.