Automatic Lecture Video Indexing Using Video OCR Technology

• DOI: 10.1109/ism.2011.26
• Publication Date: 2012-01-27T23:07:36Z
• Authors: Haojin Yang; Maria Siebert; Patrick Luhne; Harald Sack; Christoph Meinel
• Keywords: Automatic indexing; Video editing

During the last years, digital lecture libraries and video portals have become more popular. However, finding efficient methods for indexing multimedia still remains a challenging task. Since text displayed in is closely related to content, it provides valuable source retrieving contents. In this paper, we present an approach automatic based on OCR technology. We developed novel segmenter automated slide structure analysis weighted DCT (discrete cosines transformation) detector. A dynamic image constrast/brightness adaption serves purpose of enhancing quality make processible by existing common software. Time-based occurence information as well analyzed content are further used indexing. prove accuracy proposed evaluation.