Plentiful electronic gadgets for video capturing and processing attract a large number of applications using videos as their inputs. They analyze videos to gather information for their business. SBD aims to detect transition and their boundaries between consecutive shots; hence, shots with rich information are used in the content-based video indexing and retrieval. Most of the short boundary detection algorithms suffer from a large number of false positives missed and cuts. In this paper, a hybrid approach is used to reduce the false positives and ensure high coverage of cuts in the abrupt transition detection. Experiment results show that this hybrid approach has higher accuracy than individual methods such as GLCM, grey histogram and SIFT matching.
Volume 11 | 07-Special Issue
Pages: 1497-1505