A new performance evaluation software for background subtraction algorithms

In this paper, we present a new software developed to evaluate performance of background subtraction (BGS) methods in pixel-level and frame-level. To evaluate accuracy of diverse conventional BGS techniques, users first should select a BGS method, its fundamental parameters and test data set. Then, the software performs implemented analysis procedures to output qualitative and quantitative evaluation results in terms of precision, recall and f-measure. The software will be very useful in evaluating user-provided BGS methods against existing BGS methods on a variety of test data sets.