Intent Aware Optimization for Content Based Lecture Video Retrieval Using Grey Wolf Optimizer

Abstract

Nowadays, video recordings are widely used and ease to spread the knowledge among the students. Due to the rapid development of recording technologies and video based learning, the large number of videos is published in the Internet. The main challenge is to retrieve the appropriate video based on the user requirement. To achieve this objective, the newly designed intent aware optimization is proposed in this paper using grey wolf optimizer. The extraction of key frame is the initial step in the proposed system. Then, based on the input query, the keywords from the key frame are recognized by the optical character recognition and LVP. After the features are extracted, the PENN classifier is utilized to retrieve the relevant videos for the text or video query. Subsequently, the user selects one video which is used for the matching purpose based on the optimization. Thus, the intent aware optimization is newly designed using the grey wolf optimizer. The grey wolf optimizer is applied to the input database where the optimal solution is acquired by the clustering task. Finally, the user selected video is matched with the optimal solution to retrieve the more relevant video for the input query. The experimental results are validated and the performance is analysed by the parameters are F-measure, Recall and Precision. Then, the performance is also compared with the existing systems using MATLAB implementation. Thus, the higher precision value of 75% is attained which ensures the efficient retrieval of content based lecture video.