Frame Differencing, Single Gaussian, and Modified GMM for Foreground Object Detection and F-Score Measurement

Abstract

Aim- The aim of the study is to evaluate the performance of the novel Modified Gaussian mixture model algorithm in detecting foreground objects by comparing it with Frame Differencing, Single Gausian method, Gausian mixture model in four different scenarios. Materials and methods- A total of 6050 frames were taken from highway, office, pedestrians and pets2006 datasets and those video sequences have 1700, 2050, 1100 and 1200 frames respectively. Precision, Recall and F-score were calculated to evaluate the performance of all the algorithms. Results-The mean F-score values of Frame differencing, Single Gaussian, Gaussian Mixture Model and Modified GMM model algorithms are 0.4313, 0.0299, 0.5779 and 0.7379 respectively. The Modified GMM model provided the best average F-score and it is significantly better than that of remaining three models (p<0.0001, =0.05, power=80%). Conclusion- This study concluded that the novel Modified GMM algorithm performed better than the other three algorithms in these four video sequences.