Video Shot Boundary Detection Method Using Modified Artificial Bee Colony and Fast Accelerated Segment Test Algorithm
Israa Hadi, Hikmat Z. Neima, . . · International Journal of Engineering & Technology · 2018
Over recent years, video has become one of the most important means of communication. This is due to the rapid growth in network bandwidth and device storage capacity. The management of video databases therefore requires efficient video analysis methods. Video shot boundary detection (SBD) is considered an essential step in a wide range of video analysis processes such as video browsing, indexing, retrieval and summarization. Many SBD methods have been proposed; however, accuracy in the detection of shot boundaries is still unsatisfactory. In this paper, an efficient SBD method is proposed. The proposed method consists of two stages: identification and verification. In the identification stage, candidate shot boundaries are identified based on a modified Artificial Bee Colony (ABC) algorithm. In the verification stage, the global and local features of frames are extracted to verify the shot boundaries. Fuzzy histograms are utilized as a global feature while features from Fast Accelerated Segment Test (FAST) corner detection is implemented to extract interest points as a local feature. The proposed method has shown highly accurate results.