International Journal of Engineering, Management & Sciences (IJEMS) ISSN-2348 –3733, Volume-1, Issue-4, April 2014 5 www.alliedjournals.com Abstract— Motion-based features play important role since they are closely related to the ‘dynamic’ nature of videos. Motion segmentation techniques means to separate non-stationary objects from stationary backgrounds and it is analogous to separating an object from the background in image processing. Global thresholds, i.e., constant thresholds are chosen in the other methods. A common problem of global thresholding is that in practice it is impossible to find a single global threshold that works with all kinds of video materials. Choosing the right threshold is an important problem both in the color histogram comparison and edge change ratio algorithms. In this novel technique one will use the adaptive thresholding techniques through which one get the better PSNR results. Keywords—Block based video compression, Thresholding I. INTRODUCTION Video processing techniques such as video compression and video content analysis have been widely used in various applications. However, in many applications, video processing steps are often integrated with the video compression module. Video compression is the application of data compression on digital videos. The fundamental aim is to minimize redundancy of the multimedia data in order to be able to store or transmit data in an efficient form. In most images or frames the neighboring pixels are correlated and therefore contain redundant information. Three fundamental components of compression are Coding Redundancy, Spatial Redundancy and Temporal Redundancy, Irrelevant Information. There are two ways of classifying compression techniques are Lossless vs. Lossy compression and Predictive vs. Transform coding. There are many methods used for video compression, and the most famous and apply technique is Joint Picture Experts Group, which is an ISO/ITU standard for compressing digital video. JPEG performs lossy compression for each frame similar to JPEG, which means pixels from the original images are permanently removed. The grey scale image gives 256 levels of possible intensity for each pixel, so these images refer to 8 bits per pixel (bpp). The typical RGB color images, with 8 bits for Red, 8 bits for Green, and 8 bits for Blue, then the intensity I is defined by (I=R+G+B). The human eye is most sensitive to variations in intensity, so the most difficult part of compressing a color image lies in the compressing of the intensity. Digital video consists of a stream of images captured at regular time intervals. The images are represented as digitized samples containing visual (color and intensity) Manuscript received April 12, 2014. Mukhvinder Singh, Computer Science and Engineering SSCET, Pathankot, India Sachin Sharma, Computer Science and Engineering KITE, Jaipur, India Deepankur Bansal, Computer Science and Engineering ACEIT, Jaipur, India information at each spatial and temporal location. Visual information at each sample point may be represented by the values of the three basic color components RGB color space. A video signal can be sampled in either frames or fields. In this paper we are proposing a technique by which we can separate the stationary and moving objects in real time s as to result in a lossless video compression. Lossless video compression means that the compressed file after decompressing will be exactly same as the original video. Now days the techniques which are being used for video compression are all lossy compression type unlike ours “Object repetition based video compression”. In this paper we present an object repetition based video coding approach that retains the relative advantages of both the hybrid based and block-based coders while minimizing the drawbacks of both. By employing motion segmentation techniques to separate moving objects from stationary backgrounds, the coder optimizes the bit allocation to those areas that are changing most frequently. This technique also provides the ability to selectively encode, decode, and manipulate individual objects in a video stream and, hence, supports content-based functionalities such as object scalability and object manipulation easily. The quality of image is degraded by various noises in its acquisition and transmission. Image De noising has remained a fundamental problem in the field of image processing. There is various noise reduction techniques used for removing noise. Most of the standard algorithms use to de noise the noisy image and perform the individual filtering process which reduces the noise level. But the image is either blurred or over smoothed due to the loss of edges. Noise reduction is used to remove the noise without losing detail contained in the images. Edge detection and thresholding were used to locate edge areas in both the original and degraded video sequences. Degradation in the edge areas was calculated by measuring the peak signal-to noise ratio (PSNR) between the edge areas of the original and degraded video clips.MSE is a widely used full reference objective measure in modern block-based video compression algorithms such as H.264/AVC. It is employed by the rate-distortion optimized mode selection process as a quality measure for choosing the best compression option that gives an optimal tradeoff between picture quality and data rate. Compression Ratio: (1) Where n1 is the data rate of original image and n2 is that of the encoded bit-stream. Root Mean square error and peak signal-to noise ratio mathematically represented as: Thresholding Based Video Compression Mukhvinder Singh, Sachin Sharma, Deepankur Bansal 1 2  n CR n