- 上一篇：基于EAP的异构数据迁移
- 下一篇：多自由度越障机构动力学建模及路径规划
摘要数字图像抠图技术是把图像中特定部分从其他部分中分离出来的一种图像处理技术，它最初被运用于电影工业里的特效制作，为电影产业创造了可观的商业价值。随着科学技术的不断发展以及人类需求的不断提高，抠图技术也从固定背景的抠图技术发展到了自然图像的抠图技术，并且广泛运用于杂志设计，平面艺术，广告，家庭娱乐，影视后期制作等诸多领域。64534
本文的主要工作有如下4点：
2.简要介绍了几种常见的抠图算法。
3.研究了基于少量交互信息的自然图像抠图的最大相似度区域合并算法，该算法通过少量交互信息的输入，获得目标前景与背景的主要特征，从而将感兴趣目标从背景中抽取出来。
4.给出了该算法的详细设计流程与实现方案，并用软件实现了该算法。实验结果表明，在少量标记的情况下，该算法可以有效的地提取感兴趣目标，同时目标边界轮廓保留比较准确，可以用于图像编辑，合成等领域。
毕业论文关键词：数字抠图，自然图像抠图，最大相似度规则，区域合并，目标前景，背景
毕业设计说明书（论文）外文摘要
Title natural image matting and software implementation
Abstract
Digital image matting is an image processing technology extracting some certain parts(foreground) from the other parts(background). This technology is used for the special effects in film industrial, and has made great commercial value for the industrial. With development of science and technology and the increasing human demanding, image matting has evolved from fixed background image matting to natural image matting, and been widely used in magazine design, graphic arts, advertisement, family entertainment, video post-production and etc.
The following 4 works has been done in this paper:
1. Introduce the meaning of the research and the domestic and overseas current status
2. Introduce several common matting algorithm.
3. Study the region merging algorithm of interactive information based natural image matting, which can extract the interested object from the background by obtaining the main features of object and back ground from the interactive information.
4. Propose the detailed design and the implementation scheme of the algorithm, and implement it with software. According to the experiments we have done, the algorithm can extract the interested object with several markers, while preserve the integrity of the object contour, which can be used in image edit and combining.
Keywords: digital matting, natural image matting, maximal similarity rule, segmentation, region merging, object foreground, background
1 绪论 5
1.1 研究背景 5
1.2 理论基础及问题定义 6
1.3 国内外发展动态 9
1.3.1 蓝屏抠图技术 10
1.3.2 基于trimap的自然图像抠图技术 11
1.3.3 基于线条的自然图像抠图技术 11
1.4 本课题的研究工作以及结构安排 12
2 几种常见的自然图像抠图算法 12
2.1 泊松抠图算法 12
2.2 Knockout抠图算法 13
2.3 鲁棒抠图算法 14
2.4 贝叶斯抠图算法