Advanced search
1 file | 1.16 MB

Real time algorithm invariant to natural lighting with LBP techniques through an adaptive thresholding implemented in GPU processors

Author
Organization
Abstract
Video analysis in real time requires fast and efficient algorithms to extract relevant information from a considerable number, commonly 25, of frames per second. Furthermore, robust algorithms for outdoor visual scenes may retrieve correspondent features along the day where a challenge is to deal with lighting changes. Currently, Local Binary Pattern (LBP) techniques are widely used for extracting features due to their robustness to illumination changes and the low requirements for implementation. We propose to compute an automatic threshold based on the distribution of the intensity residuals resulting from the pairwise comparisons when using LBP techniques. The intensity residuals distribution can be modelled by a Generalized Gaussian Distribution (GGD). In this paper we compute the adaptive threshold using the parameters of the GGD. We present a CUDA implementation of our proposed algorithm. We use the LBPSYM technique. Our approach is tested on videos of four different urban scenes with mobilities captured during day and night. The extracted features can be used in a further step to determine patterns, identify objects or detect background. However, further research must be conducted for blurring correction since the scenes at night are commonly blurred due to artificial lighting.
Keywords
TEXTURE CLASSIFICATION, Local binary patterns, Robustness to artificial lighting, image analysis using CUDA, FEATURE DISTRIBUTIONS

Downloads

  • (...).PDF
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 1.16 MB

Citation

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

Chicago
Orjuela Vargas, Sergio Alejandro, J Triana-Martinez, JP Yañez, and Wilfried Philips. 2014. “Real Time Algorithm Invariant to Natural Lighting with LBP Techniques Through an Adaptive Thresholding Implemented in GPU Processors.” In Proceedings of SPIE. Vol. 9023.
APA
Orjuela Vargas, S. A., Triana-Martinez, J., Yañez, J., & Philips, W. (2014). Real time algorithm invariant to natural lighting with LBP techniques through an adaptive thresholding implemented in GPU processors. Proceedings of SPIE (Vol. 9023). Presented at the Conference on Digital Photography X.
Vancouver
1.
Orjuela Vargas SA, Triana-Martinez J, Yañez J, Philips W. Real time algorithm invariant to natural lighting with LBP techniques through an adaptive thresholding implemented in GPU processors. Proceedings of SPIE. 2014.
MLA
Orjuela Vargas, Sergio Alejandro, J Triana-Martinez, JP Yañez, et al. “Real Time Algorithm Invariant to Natural Lighting with LBP Techniques Through an Adaptive Thresholding Implemented in GPU Processors.” Proceedings of SPIE. Vol. 9023. 2014. Print.
@inproceedings{8027599,
  abstract     = {Video analysis in real time requires fast and efficient algorithms to extract relevant information from a considerable number, commonly 25, of frames per second. Furthermore, robust algorithms for outdoor visual scenes may retrieve correspondent features along the day where a challenge is to deal with lighting changes. Currently, Local Binary Pattern (LBP) techniques are widely used for extracting features due to their robustness to illumination changes and the low requirements for implementation. We propose to compute an automatic threshold based on the distribution of the intensity residuals resulting from the pairwise comparisons when using LBP techniques. The intensity residuals distribution can be modelled by a Generalized Gaussian Distribution (GGD). In this paper we compute the adaptive threshold using the parameters of the GGD. We present a CUDA implementation of our proposed algorithm. We use the LBPSYM technique. Our approach is tested on videos of four different urban scenes with mobilities captured during day and night. The extracted features can be used in a further step to determine patterns, identify objects or detect background. However, further research must be conducted for blurring correction since the scenes at night are commonly blurred due to artificial lighting.},
  articleno    = {902304},
  author       = {Orjuela Vargas, Sergio Alejandro and Triana-Martinez, J and Ya{\~n}ez, JP and Philips, Wilfried},
  booktitle    = {Proceedings of SPIE},
  isbn         = {978-0-8194-9940-0},
  issn         = {0277-786X},
  keyword      = {TEXTURE CLASSIFICATION,Local binary patterns,Robustness to artificial lighting,image analysis using CUDA,FEATURE DISTRIBUTIONS},
  language     = {eng},
  location     = {San Francisco, CA},
  title        = {Real time algorithm invariant to natural lighting with LBP techniques through an adaptive thresholding implemented in GPU processors},
  url          = {http://dx.doi.org/10.1117/12.2042619},
  volume       = {9023},
  year         = {2014},
}

Altmetric
View in Altmetric
Web of Science
Times cited: