Segmentation of page images using the area Voronoi diagram
Computer Vision and Image Understanding - Special issue on document image understanding and retrieval
Digital Image Processing
The Document Spectrum for Page Layout Analysis
IEEE Transactions on Pattern Analysis and Machine Intelligence
IBM Journal of Research and Development
Finding the best-fit bounding-boxes
DAS'06 Proceedings of the 7th international conference on Document Analysis Systems
Hi-index | 0.00 |
Patent document images maintained by the U.S. patent database have a specific format, in which figures and descriptions are separated into different pages. This makes it difficult for users to refer to a figure while reading the description or vice versa. The system introduced in this paper is to prepare these patent documents for a friendly browsing interface. The system is able to segment an imaged page with several figures into individual figures and extract caption and label information from the figure. After obtaining captions and labels, figures and the relevant description are linked together, and thus users could easily refer from a description to the figure or vice versa.