VIDEO SHOT BOUNDARY DETECTION BASED ON NODAL ANALYSIS OF GRAPH THEORITIC APPROACH

Extraction of video shots is a promising step in the process of shot boundary detection. Most of the existing methods measure discontinuities between the two consecutive video frames based on its low level features. In this paper we present an innovative method that take the advantage of previously define method to measure discontinuity between the frames. The conceptual knowledge of nodal analysis is combined with the existing technique: histogram difference method and statistical deviation of pixel intensities using contrast change parameters to detect the edit effect occurring in different videos. These effects include both abrupt transition and gradual transition. Proposed method is tested on different videos and the result shows its accuracy and efficiency in detecting shot boundaries.