Moving objects detection is a fundamental step in many vision based applications. Background subtraction is the typical method. When scene exhibits pertinent dynamism method based on mixture of Gaussians is a good balance between accuracy and complexity, but fails due to two kinds of false segmentations i.e moving shadows incorrectly detected as objects and some actual moving objects not detected as moving objects. In computer vision, segmentation refers to process of partitioning a digital image in to multiple segments and goal of segmentation is to simplify and/or change representation of image in to something that is more meaningful and easier to analyse. A colour clustering based on k-means and image over-segmentation are used to segment the input frame into patches and shadow suppression done by HSV colour space, the outputs of mixture of Gaussians are combined with the colour clustered regions to a module for area confidence measurement. In this way, two major segment errors can be corrected. Experimental results show that the proposed approach can significantly enhance segmentation results.
Background Subtraction Based on a New Fuzzy Mixture of Gaussians for Moving Object Detection
