A Self-Organizing Approach to Background Subtraction for Visual Surveillance Applications

By Lucia Maddalena, Alfredo Petrosino. IEEE Transactions on Image Processing, 2008

843 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/eb004bf1701faa7e3201555d3a615eb2b1429b27

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Traditional and recent approaches in background modeling for foreground detection: An overview (2014, methodology): “Self Organizing Neural Network (SONN) (10) Maddalena and Petrosino (2008) [176]”
Evaluation of background subtraction techniques for video surveillance (2011, background): “Maddalena 5 [ 19 ] 2008 multimodal (mean) color pixel 7 yes DM”
Survey on Contemporary Remote Surveillance Systems for Public Safety (2010, background): “This type of a problem is known to be essential and demanding [25].”
Shadow detection: A survey and comparative evaluation of recent methods (2012, methodology): “Traditionally, simple and fast shadow detection and removal methods have been favoured in computer vision applications such as tracking systems [9, 16, 19, 44].”
Changedetection.net: A new change detection benchmark dataset (2012, methodology): “Out of the above methods, all except for the Euclidean and Mahalanobis distance methods [4], are robust to background motion, four are robust to shadows [13, 28, 12, 18] and two are robust to artifacts stemming from intermittent motion [2, 7].”
A comprehensive review of background subtraction algorithms evaluated with synthetic and real videos (2014, methodology): “In this experiment, the LBAdaptiveSOM and AdaptiveBackgroundLearning algorithms show interesting results, since it is possible to improve their good detection without increasing false ones too much.”
Background segmentation with feedback: The Pixel-Based Adaptive Segmenter (2012, neutral): “SOBS [5] 0.”
Recent Advanced Statistical Background Modeling for Foreground Detection - A Systematic Survey (2011, background): “The network learns how to classify each pixel as backgr ound or foreground [332][333].”
SuBSENSE: A Universal Change Detection Method With Local Adaptive Sensitivity (2015, methodology): “Unprecedented methods based on artificial neural networks [21], [22] have been proposed and achieve good”
Video Processing From Electro-Optical Sensors for Object Detection and Tracking in a Maritime Environment: A Survey (2016, methodology): “These techniques are LBP [135], LBSP [172], FuzzGMM [185], FAdaSOM [183], and Eigen [162].”
Background Subtraction in Real Applications: Challenges, Current Models and Future Directions (2019, methodology): “[307] designed a spatio-Contextual GMM (SC-GMM) that outperforms 18 background subtraction algorithms such as 1) classical algorithms like the original MOG, the original KDE and the original PCA, and 2) advanced algorithms like DPGMM (VarDMM) [132], PBAS [142], PBAS-PID [356], SuBSENSE [340] and SOB”
ViBe: A Universal Background Subtraction Algorithm for Video Sequences (2011, background): “Classification algorithms are also the basis of other algorithms, as in the one provided in [15], where the background model learns its motion”