Moving object detection for visual surveillance using quasi-euclidian distance

Abstract

Moving object detection is a fundamental step for visual surveillance system, other image processing, and computer vision applications. The most popular and common technique for moving foreground detection is background subtraction. In dynamic background, Gaussian Mixture Model performs better for object detection. In this work, a GMM-based background model is developed. This work proposes a quasi-euclidian distance measure in order to measure the variation in terms of distance, between modeled frame and test frame. To classify the pixel, this distance is compared with a suitable threshold. The connected component and blob labeling has been used to improve the model with a threshold. Morphological filter is used to improve the foreground information. The experimental study shows that the proposed work performs better in comparison to considered state-of-the-art methods in term precision, recall, and f-measure.