Chap 6. Image Retrieval
• Global Feature and Local Feature
• Indexing Structure
• Relevance Feedback
• Evaluation Pipeline
Color Histogram
Advantage :
➢ The color histogram is easy to compute and effective in characterizing
both the global distribution of colors in an image.
➢ Robust to translation and rotation about the view axis and changes
only slightly with the scale, occlusion and viewing angle.
Disadvantage :
➢ Without color distributions of images
Texture Feature
• Tamura Texture Feature
• Histogram Moments
• Co-occurrence Matrix
• Fractional Brownian Motion
• Wavelet Transform
• Simultaneous Auto-regressive
• Gabor Transform
The image retrieval system is used for retrieving images related to the user request from the database. In the presented image retrieval system, the set of texture features was extracted and incorporated into the NS domain to represent image content in the training dataset (Eisa, 2014). Another technique called unsupervised learning image classification has been introduced, and it is based on the integration of optimization linear programming and neutrosophic sets (Salama, Eisa, ElGhawalby, & Fawzy, 2016). In this, texture features are presented for embedding images in the neutrosophic domain. This set of features is used for image retrieval using the neutrosophic domain.
图像检索系统用于从数据库中检索与用户请求有关的图像。在提出的图像检索系统中,提取了一组纹理特征并将其纳入NS域,以表示训练数据集中的图像内容(Eisa,2014)。另一项名为无监督学习图像分类的技术已经被引入,它是基于优化线性编程和中性集的整合(Salama, Eisa, ElGhawalby, & Fawzy, 2016)。其中,纹理特征被提出来用于嵌入中子域的图像。这组特征被用于使用中微子域的图像检索。
Chap 7. Video Retrieval
▪ 随着数据获取、存储、传输技术的快速发展,人们可以轻易地查询、获取和产生大量的视频信息
▪ 描述、组织和管理视频数据的工具和技术十分有限
▪ 1. 视频数据提供的信息量大▪ 2. 视频数据之间关系复杂▪ 可以看做是文本、音频以及含时间维度的图像的集合▪ 3. 视频数据解释的多样性与模糊性
▪ 视频索引和检索的方法包括:
▪ 1. 基于元数据的方法▪ 元数据包括标题、作者、导演等
▪ 2. 基于文本的方法▪ 例如视频中的字幕文本
▪ 3. 基于音频的方法
▪ 4. 基于内容的方法▪ 基于帧的检索▪ 基于镜头—关键帧的检索
▪ 基于内容的视频检索
▪ 根据视频的内容和上下文关系,在没有人工参与的情况下,自动提取并描述视频的特征和内容,在此基础上进行检索
▪ 与文档类似,视频也需要结构化