[Show abstract][Hide abstract] ABSTRACT: This paper presents the application of multi dimensional feature reduction ofConsistency Subset Evaluator (CSE) and Principal Component Analysis (PCA)and Unsupervised Expectation Maximization (UEM) classifier for imagingsurveillance system. Recently, research in image processing has raised muchinterest in the security surveillance systems community. Weapon detection is oneof the greatest challenges facing by the community recently. In order toovercome this issue, application of the UEM classifier is performed to focus onthe need of detecting dangerous weapons. However, CSE and PCA are used toexplore the usefulness of each feature and reduce the multi dimensional featuresto simplified features with no underlying hidden structure. In this paper, we takeadvantage of the simplified features and classifier to categorize images objectwith the hope to detect dangerous weapons effectively. In order to validate theeffectiveness of the UEM classifier, several classifiers are used to compare theoverall accuracy of the system with the compliment from the features reduction ofCSE and PCA. These unsupervised classifiers include Farthest First, DensitybasedClustering and k-Means methods. The final outcome of this researchclearly indicates that UEM has the ability in improving the classification accuracyusing the extracted features from the multi-dimensional feature reduction of CSE.Besides, it is also shown that PCA is able to speed-up the computational timewith the reduced dimensionality of the features compromising the slight decreaseof accuracy.