融合多特征的运动一致性图像分割

目的:在彩色图像分割中，光流法能够得到运动区域，但难以获得运动目标准确的分割边界，而常用的算法往往会产生过分割。为了克服光流法的不足，在保留显著性区域的同时抑制过分割，从而获得具有运动一致性区域的分割结果，提出融合多特征的运动一致性图像分割算法。方法:首先通过Mean Shift算法获取图像的初始分割，然后利用空域信息（包括颜色、边缘和区域面积）对视觉感知上具有相似性的区域进行合并，再利用时域信息进行运动一致性区域合并，最终得到分割结果。结果:实验结果表明通过结合时空信息，该方法能够有效抑制过分割，不仅弥补了光流场不能准确提取目标边缘的不足，而且提高了分割目标的完整性。结论:与两种流行的彩色图像分割算法相比，所提方法获得了更加理想的结果。

Motion coherence image segmentation fused with multi-feature

Objective: During the segmentation of color image, optical flow methods can acquire the moving regions, but it can hardly obtain the correct segmentation boundaries of moving objects, meanwhile, familiar algorithms usually suffer over-segmentation. To overcome the shortage of optical flow method and suppress the over-segmentation, meanwhile, preserve the salient regions, a new motion coherence image segmentation algorithm fused with multi-feature is proposed. Method: Firstly, the initial regions are acquired by Mean Shift algorithm; Secondly, the regions with homogenization of visual sensing are merged by utilizing spatial information (including color, edge and region area); Thirdly, the motion coherence regions are merged by using temporal information; Finally, the segmentation result is obtained. Result: The experimental results demonstrate that by combining spatial-temporal information, the proposed method can suppress over-segmentation effectively, not only does it make up the shortage that correct object edges can hardly acquire by using optical flow, but increases the completeness of segmented objects. Conclusion: Compared with two popular color image segmentation algorithms, our method gets more ideal results.