Knowledge-Based Image Retrieval with Spatial and Temporal Constructs

本研究项目采用基于知识的方法，通过具有空间和时间结构的特征和内容来检索医学图像。对医学图像中的所选感兴趣对象(例如，X射线、MR图像)进行分割，并从这些对象生成轮廓。特征(例如，形状、大小)和内容(例如，对象之间的空间关系)被提取并存储在特征和内容数据库中。关于图像特征的知识可以表示为称为类型抽象层次(TAH)的层次结构。提出了一种基于知识的图像语义模型，该模型由原始数据层、特征和内容层、模式层和知识层四个层次组成，用于表示图像对象特征的各个方面。该模型提供了一种访问和处理空间、演化和时间查询的机制。形成了一种基于知识的时空查询语言(KSTL)，该语言支持通过特征和内容、概念术语和时态逻辑谓词进行近似匹配的算子。此外，还开发了一种可视化查询语言，它接受屏幕上映射到KSTL的点击和拖动可视图标输入。本研究的结果适用于各种多媒体信息系统。

This research project employs a knowledge-based approach to retrieve medical images by feature and content with spatial and temporal constructs. Selected objects of interest in a medical image (e.g. x-ray, MR image) are segmented, and contours are generated from these objects. Features (e.g., shape, size) and content (e.g., spatial relationships among objects) are extracted and stored in a feature and content database. Knowledge about image features can be expressed as a hierarchical structure called a Type Abstraction Hierarchy (TAH). A knowledge-based semantic image model is developed which consists of four layers (raw data layer, feature and content layer, schema layer, and knowledge layer) to represent the various aspects of an image objects' characteristics. The model provides a mechanism for accessing and processing spatial, evolutionary, and temporal queries. A knowledge-based spatial temporal query language (KSTL) is formed which supports operators for approximate matching via feature and content, conceptual terms, and temporal logic predicates. Further, a visual query language is also developed that accepts point-click-and-drag visual iconic input on the screen that maps into KSTL. The results from this research are applicable to a variety of multimedia information systems.