Advanced search
1 file | 720.14 KB

Intelligent pre-processing for fast-moving object detection

Chris Poppe (UGent) , Sarah De Bruyne (UGent) , Gaëtan Martens (UGent) , Peter Lambert (UGent) and Rik Van de Walle (UGent)
Author
Organization
Abstract
Detection and segmentation of objects of interest in image sequences is the first major processing step in visual surveillance applications. The outcome is used for further processing, such as object tracking, interpretation, and classification of objects and their trajectories. To speed up the algorithms for moving object detection, many applications use techniques such as frame rate reduction. However, temporal consistency is an important feature in the analysis of surveillance video, especially for tracking objects. Another technique is the downscaling of the images before analysis, after which the images are up-sampled to regain the original size. This method, however, increases the effect of false detections. We propose a different pre-processing step in which we use a checkerboard-like mask to decide which pixels to process. For each frame the mask is inverted to avoid that certain pixel positions are never analyzed. In a post-processing step we use spatial interpolation to predict the detection results for the pixels which were not analyzed. To evaluate our system we have combined it with a background subtraction technique based on a mixture of Gaussian models. Results show that the models do not get corrupted by using our mask and we can reduce the processing time with over 45% while achieving similar detection results as the conventional technique.
Keywords
Moving object detection, Mixture of Gaussian Models, video surveillance

Downloads

  • 2008.03 - SPIE Defense and Security - Chris Poppe et al. - Intelligent Pre-processing for Fast Moving Object Detection.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 720.14 KB

Citation

Please use this url to cite or link to this publication:

Chicago
Poppe, Chris, Sarah De Bruyne, Gaëtan Martens, Peter Lambert, and Rik Van de Walle. 2008. “Intelligent Pre-processing for Fast-moving Object Detection.” In Proceedings of the Society of Photo-optical Instrumentation Engineers (spie), ed. Z Rahman, SE Reichenbach, and MA Neifeld. Vol. 6978. Bellingham, WA, USA: SPIE, the International Society for Optical Engineering.
APA
Poppe, Chris, De Bruyne, S., Martens, G., Lambert, P., & Van de Walle, R. (2008). Intelligent pre-processing for fast-moving object detection. In Z Rahman, S. Reichenbach, & M. Neifeld (Eds.), PROCEEDINGS OF THE SOCIETY OF PHOTO-OPTICAL INSTRUMENTATION ENGINEERS (SPIE) (Vol. 6978). Presented at the Conference on Visual Information Processing XVII, Bellingham, WA, USA: SPIE, the International Society for Optical Engineering.
Vancouver
1.
Poppe C, De Bruyne S, Martens G, Lambert P, Van de Walle R. Intelligent pre-processing for fast-moving object detection. In: Rahman Z, Reichenbach S, Neifeld M, editors. PROCEEDINGS OF THE SOCIETY OF PHOTO-OPTICAL INSTRUMENTATION ENGINEERS (SPIE). Bellingham, WA, USA: SPIE, the International Society for Optical Engineering; 2008.
MLA
Poppe, Chris, Sarah De Bruyne, Gaëtan Martens, et al. “Intelligent Pre-processing for Fast-moving Object Detection.” Proceedings of the Society of Photo-optical Instrumentation Engineers (spie). Ed. Z Rahman, SE Reichenbach, & MA Neifeld. Vol. 6978. Bellingham, WA, USA: SPIE, the International Society for Optical Engineering, 2008. Print.
@inproceedings{828648,
  abstract     = {Detection and segmentation of objects of interest in image sequences is the first major processing step in visual surveillance applications. The outcome is used for further processing, such as object tracking, interpretation, and classification of objects and their trajectories. To speed up the algorithms for moving object detection, many applications use techniques such as frame rate reduction. However, temporal consistency is an important feature in the analysis of surveillance video, especially for tracking objects. Another technique is the downscaling of the images before analysis, after which the images are up-sampled to regain the original size. This method, however, increases the effect of false detections. We propose a different pre-processing step in which we use a checkerboard-like mask to decide which pixels to process. For each frame the mask is inverted to avoid that certain pixel positions are never analyzed. In a post-processing step we use spatial interpolation to predict the detection results for the pixels which were not analyzed. To evaluate our system we have combined it with a background subtraction technique based on a mixture of Gaussian models. Results show that the models do not get corrupted by using our mask and we can reduce the processing time with over 45\% while achieving similar detection results as the conventional technique.},
  author       = {Poppe, Chris and De Bruyne, Sarah and Martens, Ga{\"e}tan and Lambert, Peter and Van de Walle, Rik},
  booktitle    = {PROCEEDINGS OF THE SOCIETY OF PHOTO-OPTICAL INSTRUMENTATION ENGINEERS (SPIE)},
  editor       = {Rahman, Z and Reichenbach, SE and Neifeld, MA},
  isbn         = {0277-786X},
  language     = {eng},
  location     = {Orlando, FL, USA},
  publisher    = {SPIE, the International Society for Optical Engineering},
  title        = {Intelligent pre-processing for fast-moving object detection},
  url          = {http://dx.doi.org/10.1117/12.777034},
  volume       = {6978},
  year         = {2008},
}

Altmetric
View in Altmetric
Web of Science
Times cited: