European Journal of Molecular & Clinical Medicine

Abstract

Due to fast expansion in computer and telecommunication industry digital video has become a promising force for its large data storage. Nowadays lecture video and audio data is growing fast and becoming enlarged in internet. Therefore a Video Retrieval and Summarization with effective Key Frame Extraction for content based lecture video search and indexing is presented here by introducing video segmentation and text key frame extraction. Video contents can be extracted through detection of metadata. By applying automated speech recognition and optical character recognition the metadata from the lecture video can be mined. Key frame representation is a powerful mechanism for accessing the video content and key frame extraction is really helpful in retrieving the accurate video content during browsing. Transition Outcome Recognition (TOR) method is proposed to automatically segment the video streams into shots. Optical Character Recognition (OCR) process is used to extract the text key-frames that reduces frame redundancy and captures slide transition in the video shot. Finally, a self-concentration replica is introduced to select key-frames sequences inside shots, thus key static images are selected as video content summarization.