Abstract: Motion Detection is a process of detecting a change in the position of the object relative to its environment or vice-versa. As time changes computer applications are becoming important part in every field. Due to which it has been used in several applications. The verification of the object from a video and then tracking of that object is an significant task in computer vision. Object is a thing of interest whose movement is to record. It can be anything of any dimensions. Video is a recording of moving visual images and to record the movements of the object in the video is quite a complex task. Object detection is termed as to detect or locate objects from consecutive frames of a video file. For object detection, object representation is required. There are many techniques that came into existence to capture the motions of the object but every technique is having their own merits and demerits. One of the merit is that they capture every motion of the object in the video while demerit contains the concept of Computational Cost, Accuracy, Time, Noise, Shadow Effect, only Major Movement Detection, Detection of Stationary Objects ,etc. Also most of the cameras produce a noisy image, which result into a motion in that such places, where there is no motion at all. This paper presents a survey of different motion detection techniques. This survey paper includes background subtraction method, temporal differencing, statistical approach, and optical flow method and then comparison is made on them, with this a feature extraction algorithm can be used to examine the difference in the frames and thus, object can be detected in it. The purpose of a feature extraction is to obtain descriptive quantities (description) and reduce the dimensionality of data without losing relevant information. The dimensionality of a data defines to the number of values (i.e. dimensions) of a single measurement. A visual feature refers here are: - colour, texture and shape, etc.
Keywords: Video Tracking; Motion Detection,Feature Extraction,Dimensionality.