Multiple Objects Detection Based on Background Subtraction and Gaussian Pyramid
Liang Jiang, Yongzhao Zhan · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2013
For the multi-object detection problems with complex background, a method which combines background subtraction with Guassian pyramid on objects detection is presented.The method detects the whole object with taking samples from the objects with Guassian pyramid, building background models, extracting foreground areas from background subtraction, and eliminating the shadow on the foreground.The detection that integrates Gauss model which concerns the background renewal of calculation, overcomes the error resulted from the sudden change of background.A dynamic threshold concept is proposed to enhance detection effect, thus it increases the possibility of implementation.The experiment results show that moving objects detected by the proposed method is more reliable than the state-of-the-art.