Abstract
Background subtraction without a separate training phase has become a critical task, because a sufficiently long and clean training sequence is usually unavailable, and people generally thirst for immediate detection results from the first frame of a video. Without a training phase, we propose a background subtraction method based on three-dimensional (3D) discrete wavelet transform (DWT). Static backgrounds with few variations along the time axis are characterized by intensity temporal consistency in the 3D space-time domain and, hence, correspond to low-frequency components in the 3D frequency domain. Enlightened by this, we eliminate low-frequency components that correspond to static backgrounds using the 3D DWT in order to extract moving objects. Owing to the multiscale analysis property of the 3D DWT, the elimination of low-frequency components in sub-bands of the 3D DWT is equivalent to performing a pyramidal 3D filter. This 3D filter brings advantages to our method in reserving the inner parts of detected objects and reducing the ringing around object boundaries. Moreover, we make use of wavelet shrinkage to remove disturbance of intensity temporal consistency and introduce an adaptive threshold based on the entropy of the histogram to obtain optimal detection results. Experimental results show that our method works effectively in situations lacking training opportunities and outperforms several popular techniques.
Highlights
Smart video surveillance systems are extensively applied to various indoor and outdoor scenes nowadays, owing to rapidly increasing demands of security protection, healthcare, home care, etc.Moving object detection is a fundamental task of smart video surveillance [1] and has become a hot issue over the last decade [2,3,4,5,6,7,8]
We propose a background subtraction method without any training phase, based on three-dimensional (3D) discrete wavelet transform (DWT)
This will lead to many ghosts in the detection results of crowded scenes, especially for short clips
Summary
Smart video surveillance systems are extensively applied to various indoor and outdoor scenes nowadays, owing to rapidly increasing demands of security protection, healthcare, home care, etc.Moving object detection is a fundamental task of smart video surveillance [1] and has become a hot issue over the last decade [2,3,4,5,6,7,8]. Background subtraction techniques [9,10,11,12,13,14,15,16,17,18,19] are the most popular in moving object detection. Continuous moving objects in the crowded scenes (such as airports, train stations, shopping centers and buffet restaurants) make it hard to get clean training data. Researchers have proven that a much longer training phase (up to 800 frames) has to be used to build accurate background models for the crowded scenes [23]. Analysis of Background Removal in the 3D Wavelet Domain. We further detail the rationality of our method to remove the static backgrounds in the 3D wavelet domain by taking advantage of multiscale analysis characteristics of DWT. We draw a line t1 , parallel to the t axis, passing through an arbitrary point (a,b,t0 )
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
