Extracting Descriptive Frames from Informational Videos

Informational videos are becoming increasingly important among all video types. The users spend so much time browsing the informative videos, even if they not interested in their topics. Thence, a new method for extracting descriptive frames is presented this paper that allows to navigate directly topics of interest video. proposed consists three main phases: preprocessing, segmentation, and separation phase. Firstly, extracted from resized, converted grayscale. Then, divided into blocks, kurtosis moment calculated each block. segmented based on an examination differences between features moment. Lastly, grouped separate after distinguished uninformative ones using clustering technique. results demonstrated functional effectiveness method. According accuracy F1-Score measures, it has performance up 100%. Moreover, significantly summarized by reducing duration less than 1% its original time.