Abstract
This paper addresses the problem of lower precision in automatic video classification. A novel rule-based video classification approach is proposed. Firstly, in the video segmentation process, a set of video attributes is extracted to represent the content of video and a video attribute database is generated. Then the decision tree and class association rule mining techniques are performed on this video attribute database to extract a decision tree rule set and a class association rule set respectively. Lastly, a combination and pruning algorithm are applied to combine these two rule sets to generate a final classification rule set. The experimental result verifies the consistency of decision tree classification with class association classification. The result also shows the final combined rule set has higher classification precision that just one rule set.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call