Matlab基于全局对比度的视觉注意模型的比较与研究
毕业论文关键词: 全局对比度；视觉显著性；显著性检测
Research on Visual Attention Models based upon Global Contrast
Abstract: Computer image or video properly understood by the surrounding environment is the primary goal of computer vision research, the field work that is essential to the correct perception of the three main objects in the scene, the outline of object recognition and access to environmental context object. To achieve this goal, to obtain an image of the visual information is the most significant step basis. Significant visual image information to reflect the extent of the human visual system to stimulate different areas of the image. Significant image region detection techniques may be implemented to extract significant information in the image. Fast and reliable detection of significant content can be perceived image editing, image segmentation, image retrieval application provides valuable reference information; to ease barriers image content and image understanding between low-level features, making the higher level of image understanding as possible. Depth study of significant detection technology, to improve the performance of image analysis and image understanding systems, and improve the application level image processing technology has far-reaching significance.
Keywords: Global contrast ; Visual saliency; Saliency detection
目录
摘要: i
Abstract: i
1 绪论 1
1.1 研究背景及意义 1
1.2 研究现状 2
1.3 图像显著性检测方法面临的挑战 3
2. 相关技术的介绍和分析 5
2.1 显著性检测的测评方法 6
2.2 典型的显著性检测方法 7
2.3.1 LC方法 7
2.3.2 HC方法 8
2.3.3 SR方法 9
2.3.4 FT方法 10
3 系统设计 12
3.1界面设计 12
3.2 界面的激活与回调函数的生成 12
3.3 结果的生成与分析 13
3.4算法的比较 17
3.4本章小结 20
4 结论 21
致谢 22
参考文献 23
1 绪论
1.1 研究背景及意义
随着计算机性能和功能的发展，人们越来越希望计算机可以更加自主更加智能的完成任务。要实现这个目标，需要计算机能够理解周围的环境。人类感知外界信息最主要方式是通过视觉，所以计算机理解周围环境的关键是具有视觉感知处理能力。人类对外界环境的视觉感知能力经过了长时间自然选择进化到了一个很高级的水平。人类能够快速高效的分析周围环境，定位感兴趣的物体，并作出相应的反应。