The availability of easy to use video editing software has made it easy for cyber criminals to combine different videos from different sources using blue screen composition technology. This, makes the authenticity of such digital videos questionable and needs to be verified especially in the court of law. Blue Screen Composition is one of the ways to carry out video forgery using simple to use and affordable video editing software. Detecting this type of video forgery aims at revealing and observing the facts about a video so as to conclude whether the contents of the video have undergone any unethical manipulation. In this work, we propose an enhanced 3-stage foreground algorithm to detect Blue Screen manipulation in digital video. The proposed enhanced detection technique contains three (3) phases, extraction, detection and tracking. In the extraction phase, a Gaussian Mixture Model (GMM) is used to extract foreground element from a target video. Entropy function as a descriptive feature of image is extracted and calculated from the target video in the detection phase. The tracking phase seeks to use Minimum Output Sum of Squared Error (MOSSE) object tracking algorithm to fast track forged blocks of small sizes in a digital video. The result of the experiments demonstrates that the proposed detection technique can adequately detect Blue Screen video forgery when the forged region is small with a true positive detection rate of 98.02% and a false positive detection rate of 1.99%. The result of this our research can be used to
