Advanced technologies and the development of internet sources drag the attention of the researchers mainly, in the areas of video retrieval. Though there are numerous methods for video retrieval, numerous challenges exists mainly, to retrieve the videos as per the user requirements. This paper proposes an enhanced optimization algorithm, named Adaptive-Grey wolf optimizer (Adaptive-GWO) to retrieve the lecture videos. The proposed Adaptive-GWO is developed by incorporating self-adaptive concept in Grey wolf optimizer (GWO). Initially, the keyframe extraction is performed in which the keywords from the key frame are identified using optical character recognition (OCR) and Local Vector Pattern (LVP). Once the features are extracted, the Probability Extended Nearest Neighbour (PENN) is used for retrieving the relevant videos based on the text or video queries. Here, the Adaptive-GWO is applied for enabling the optimal clustering. Lastly, the video selected by the user is matched with the optimal solution to retrieve more relevant video for the input query. The proposed Adaptive-GWO outperformed other existing methods with maximal precision, recall, and F-measure with values 85.560%, 81.327% and 81.563% respectively.
Volume 11 | Issue 6
Pages: 72-82