Design Prognostic Framework for Scene Classification in Video Processing

Abstract

This paper provides the technique over selecting phenomenal frames extraction structure using SBD, KFE, and Semantic mechanism which quite easy to extract the valuable scene from the entire video. The proposed system will concentrates on findings the accuracy, precision and recall measure of inputted video from TRACEVID dataset. Here the proposed system signifies the use of thresholding technique and TBD of individual frame and calculate the various parameters like standard deviation, Mean etc. In order to legalize this claim, content based video reclamation systems were furnished using color histogram, features extraction and different approaches are applied for the supervision of the semantic temperament of each frame in the video. Also it gives major idea about the classification of the scene and algorithm which generates significant result.