点击次数：

论文类型：会议论文

发表时间：2006-06-21

收录刊物：EI、CPCI-S、Scopus

卷号：2

页面范围：COVER3-COVER3

关键字：local maximum component; component-labeling algorithm; parallelizable; layout analysis

摘要：A definition of the local maximum component (the component for short) is presented for layout analysis in document image analysis (DIA), and a novel algorithm for component labeling was described. This algorithm uses a contour tracing technique to detect and label the external contour of each component, and removes the interior area of each component from the copy of the source image. Labeling and removing are completed in a single pass over source image. Experiments on various kinds of images (title, text, picture, table and formula) show that the new definition and algorithm are more efficient and flexible than the traditional labeling ones.