Abstract: Building a scalable video on demand (VOD) architecture is essential for the optimization of the VOD service cost and to support a very large VOD content library. The peer-to-peer (P2P) VOD architecture works towards achieving this goal, by distributing the video files to the user's set-top boxes at the network edge. The details of such system, however, including numerous operational issues remain to be resolved. This paper investigates the network bandwidth requirements of a peer to peer VOD service architectures system. A network mathematical model is developed to analyze the key-parameters that have influence on the P2P network bandwidth requirement. A comparison of a centralized system to the (P2P) VOD architecture bandwidth requirements as a function of video service multicast factor and files requests rate is also presented. The research results present a systematic study on the bandwidth provisioning in P2P VOD applications.

Abstract: Camera shake causes images to be blurry. There are many techniques to de-blur an image. The process of removing image blur is called de-convolution. There are many different techniques for performing the de-convolution of images. In this paper, we propose a parallel de-blur algorithm inspired by Fergus et al. in [1]. The algorithm is performed by first estimating the blur kernel, then de-convolving the blurred image with that kernel in order to obtain the original clear image. The proposed parallel algorithm has a run time of O(log N). The algorithm is verified and tested.

Abstract: Of the problem areas, the domain of matching complex objects has received, by far, the most attention. This research work is concerned with the specific class of complicated objects, i.e. logo. The progress, particularly in this field, is still at extensive research work level, due to infinite varieties of shapes and classes which are used. Essentially, the algorithm proposed is based on Principle Component Analysis (PCA) approach. In this technique, the PCA is used to extract the features, kept inherent in the normalized pattern for later matching process. In the matching process, the extracted features of unknown pattern are mapped onto formulated feature spaces, and the distance between the source symbol and modeled symbols is used as a decision making tool. The true matching rate of 100% has been achieved for hundreds tested logos.