Video indexing through human face images using LGFA and window technique

Adaptive video monitoring settings have been extensively deployed in recent years. Smart video monitoring technology enables the acquisition and analysis of movies from various devices, as well as automatic analysis based on knowledge gathering. However, the storage capacity is restricted, and the important frames from the movie cannot be saved. Though, if the movie employs face as keyframes, it creates space and time complexity. To address this issue, the Viola-Jones Algorithm was used to detect faces from extracted keyframes in Video Indexing through Human Face Images using LGFA and the sliding window technique. The image gradient for brightness is created by integrating the sliding windowing method with LGFA, and scanning the input image horizontally takes up 70% of the facial image. As a result, Barcode as an index using the sequence table of the EAN 8 approach converts a video’s human face into an EAN-8 linear video indexing barcode and thereby reducing bandwidth, storage space, and time complexity. Regular TV series video datasets, datasets of YouTube faces, and data sets of Hollywood clips were used to evaluate the proposed technique, and shown to be effective for indexing videos based on human faces.