IOSR Journal of Computer Engineering (IOSR-JCE) e-ISSN: 2278-0661,p-ISSN: 2278-8727, Volume 19, Issue 6, Ver. IV (Nov.- Dec. 2017), PP 07-11 www.iosrjournals.org DOI: 10.9790/0661-1906040711 www.iosrjournals.org 7 | Page Video Shot Cut Detection Using Statistical Block Based Method Zehra Karhan 1 , Musa Faruk Çakır 2 , Mustafa Karhan 3 , Fatih Issı 4 1 Computer Engineering Department, OndokuzMayıs University Samsun/Turkey 234 Electronics and Automation Department, CankiriKaratekin University Cankiri/Turkey Corresponding Author:Zehra Karhan1 Abstract :In this study, a method for retrieving desired frame has been proposed in short, plain, fast and without missing part from video.The number of videos is increasing due to increased multimedia tools, social media and advancing technology.It is difficult to retrieve the request because of the increase in numbers.Video shot cut detection is focused to make this a little easier.Fast and accurate determination of video shot cut detection is an important step.As known, the videos consist of frames.Video shot cut detection was determined using the differences between the frames which constitute the video.Block-based method aims to process the images in blocks for determining of the differences between frames.In addition, it is based on the fact that statistical information is extracted and compared with the next frame.By the addition of statistical information, the difference between the frames has become more sensitive to the block-based method of detection.Using this method, desired frame is acquired quickly and sensitively from large size videos. This method also provides advantages in terms of time and recording.The application of the proposed method was implemented on the Weizmann video dataset. Keywords: shot cut detection, video, block-based, statistic, feature extraction. --------------------------------------------------------------------------------------------------------------------------------------- Date of Submission: 16-12-2017 Date of acceptance: 28-12-2017 --------------------------------------------------------------------------------------------------------------------------------------- I. Introduction The number of multimedia such as image and video is increasing by the rapid development of technology. There are also some challenges in the increase of their quantity. In addition, multimedia data stored in computer environment takes up most of the disk space (audio, video). The biggest difficultyof them is that people can not reachwhat they want from among such large amounts of data within a certain amount of time. Choosing the required one from a dataset is a very difficult process. Especially it is more difficult to do this on the video and perform it manually. At this point, it is a great advantage in terms of time as well as facilitating the work of detecting transitions by splitting a video and making a general judgement about the video.This process is called video shot detection.Shot detection studies; a field that is becoming increasingly widespread and is constantly being developed.In this area, Liu Fang and colleagues presented an article called Enhancement of video shot boundary detection using HSV color space and image subsampling. In this article, HSV color space is suggested to have developed this method in this application area to use RGB more useful and sub- samples on the image [1]. Other works in video shot detection area; in 2014 by Ravi Mishra and his colleagues have studied on the comparison of detection algorithms and dual tree complex wavelet transform for shot detection in videos. Analysis and verification of video summarization using shot boundary detection is researched by Sowmya R. in 2013. Fast video shot boundary detection based on SVD and pattern matching has been studied by Zhe Ming and Friends. Goran J. Zajic and colleagues performed a video shot boundary detection based on multifractal analysis [2-5]. Donate and his colleagues in 2010 have suggested using Shot Detection in Three-Dimensional Tracking Videos to track this video sequence over time to determine the features that make a video stand out [6]. In 2010, LihongXun and colleagues have presented a novel shot detection algorithm based on clustering. In this approach, they first propose some conclusions according to color information and then define the differences of video frames. According to these defined differences, different sub-clusters separated by k-means algorithm [7]. In this study, it became possible to detect video shot cut by looking at some statistic information and block-based method for image. It makes it possible to detectvideo shot cut automatically by evaluating these parameters.Shot detection process is performed by utilizing the differences between the frames. Block-based method with usingof statistical information has become more sensitive to calculate the difference between the frames.Thus, probability of missed shot images is low and information loss is prevented.