中文摘要：

目前,单一图像在人工交互的条件下已能实现准确分割,但如果图像数量大,则需要大量劳动量.而自动分割难以在多图像间构建统一的目标模型.为此,提出一种多视角图像的少量交互分割算法,即仅对其中一幅进行交互,对其他图像进行前景目标一致的自动分割.首先通过构建特征点轨迹图模型,建立其他图像特征点与交互图像特征点之间的连接;然后利用最短路径算法进行特征点的前景/背景分类,获得多图像间的局部对应;最后将已分类特征点作为形状先验,使用测地星凸分割算法实现多视角图像的一致分割.命中率和误报率的比较实验结果表明,文中的算法具有较好的一致分割能力,而且这种局部对应的方式还可以实现多目标的分割.

英文摘要：

With the help of human-computer interaction(HCI),a monocular image can be segmented accurately.However,manual workload of HCI is remarkably heavy if a number of images are pending processing.Furthermore,it is difficult to construct a unified model that is universally applicable to different images.Wepropose an accurate segmentation algorithm for multi-view serial images with less HCI work.HCI is performedonly on one serial image,and then the objects in the foreground of the other images can be segmentedautomatically.Based on the homography of the feature points built with a graphic model,the featurepoints in the HCI image and their corresponding points in other serial images are accordingly linked.Theshortest path algorithm classifies the feature points into foreground or background,and searches the correspondencebetween the local features of serial images.Finally,the classified feature points are used as a prioriknowledge of geodesic-star-convexity segmentation algorithm to accurately segment the serial imagestogether.The comparison of hit rate and false alarm rate con-firms the accuracy and coherence of the proposedsegmentation algorithm.Moreover,the local correspondence between the serial images can be usedfor multi-object segmentation.