Your browser does not support JavaScript!
http://iet.metastore.ingenta.com
1887

access icon openaccess Combining HWEBING and HOG-MLBP features for pedestrian detection

  • Author(s): Ling Xiao 1, 2 ; Yongjun Zhang 1, 2 ; Jian Zhang 3 ; Qian Wang 1, 2 ; Yuewei Li 1, 2
    • View affiliations
  • Source: Volume 2018, Issue 16, November 2018, p. 1421 – 1426
    DOI:  10.1049/joe.2018.8308 , Online ISSN 2051-3305
This is an open access article published by the IET under the Creative Commons Attribution-NonCommercial-NoDerivs License (http://creativecommons.org/licenses/by-nc-nd/3.0/)
Received 18/07/2018, Accepted 26/07/2018, Published 16/08/2018

Pedestrian detection has vital value in many areas such as driver assistance systems, driverless cars, intelligent tourism systems etc., but there are some difficulties that need to be solved. The algorithm with high detection rate is complex and requires substantial time. Therefore, how to improve the detection accuracy and speed has become the key of pedestrian detection. For these reasons, firstly, an improved algorithm, called hash and window enhancement of binarised normed gradients (HWEBING), based on binarised normed gradients feature is proposed. Subsequently, the authors present an improved local texture feature, namely mean of local binary pattern (MLBP), based on uniform pattern local binary pattern (ULBP) for increasing the detection rate. Finally, after using the HWEBING algorithm to get the candidate windows, the combination of MLBP feature and histograms of oriented gradients feature is extracted from these windows to further enhance the detection accuracy. Experimental results reveal that speed of using the HWEBING algorithm for pre-detection is 5.5 times faster than the traditional method of pedestrian detection. Furthermore, the detection rate of MLBP feature is 3.5 and 2.1% higher than those of ULBP and basic pattern local binary pattern (Basic-LBP), respectively.

Inspec keywords: feature extraction; image texture; object detection; pedestrians

Other keywords: driver assistance systems; HWEBING algorithm; pedestrian detection; mean of local binary pattern; Basic-LBP; intelligent tourism systems; uniform pattern local binary pattern; basic pattern local binary pattern; MLBP feature; HOG-MLBP features; texture feature; pre-detection; detection accuracy; hash and window enhancement of binarised normed gradients

Subjects: Computer vision and image processing techniques; Optical, image and video signal processing; Traffic engineering computing

References

    1. 1)
      • 1. Zhong, Z., Zhang, B., Lu, G., et al: ‘An adaptive background modeling method for foreground segmentation’, IEEE Trans. Intell. Transp. Syst., 2017, 18, (5), pp. 1109–1121.
    2. 2)
      • 11. Papageorgiou, C., Poggio, T.: ‘A trainable system for object detection’, Int. J. Comput. Vis., 2000, 38, (1), pp. 15–33.
    3. 3)
      • 10. Heikkilä, M., Pietikäinen, M., Schmid, C.: ‘Description of interest regions with local binary patterns’, Pattern Recognit., 2009, 42, (3), pp. 425–436.
    4. 4)
      • 16. Benenson, R., Mathias, M., Timofte, R., et al: ‘Pedestrian detection at 100 frames per second’. Proc. IEEE Conf. Computer Vision and Pattern Recognition, Providence, USA, June 2012, pp. 2903–2910.
    5. 5)
      • 12. Walk, S., Majer, N., Schindler, K., et al: ‘New features and insights for pedestrian detection’. Proc. IEEE Computer Society Conf. Computer Vision and Pattern Recognition, San Francisco, USA, June 2010, pp. 1030–1037.
    6. 6)
      • 2. Yang, T., Jing, L.I., Pan, Q., et al: ‘Scene modeling and statistical learning based robust pedestrian detection algorithm’, Acta Autom. Sin., 2010, 36, (4), pp. 499–508.
    7. 7)
      • 17. Cheng, M.M., Zhang, Z., Lin, W.Y., et al: ‘Bing: binarized normed gradients for objectness estimation at 300fps’. Proc. IEEE Conf. Computer Vision and Pattern Recognition, Columbus, USA, June 2014, pp. 3286–3293.
    8. 8)
      • 4. Li, W., Dai, D., Tan, M., et al: ‘Fast algorithms for linear and Kernel SVM + ’. Proc. IEEE Conf. Computer Vision and Pattern Recognition, Las Vegas, USA, June 2016, pp. 2258–2266.
    9. 9)
      • 6. Wu, J.: ‘Efficient HIKSVM learning for image classification’, IEEE Trans. Image Process. Publ. IEEE Signal Process. Soc., 2012, 21, (10), pp. 4442–4453.
    10. 10)
      • 13. Dollár, P., Tu, Z., Perona, P., et al: ‘Integral channel features’. Proc. British Machine Vision Conf., London, UK, September 2009, pp. 1–11.
    11. 11)
      • 18. Mao, J., Xiao, T., Jiang, Y., et al: ‘What can help pedestrian detection?’. Proc. IEEE Conf. Computer Vision Pattern Recognition, Honolulu, USA, July 2017, pp. 6034–6043.
    12. 12)
      • 20. Chi, J.T., Chi, E.C., Baraniuk, R.G.: ‘K-pod: a method for k-means clustering of missing data’, Am. Stat., 2016, 70, (1), pp. 91–99.
    13. 13)
      • 19. Redmon, J., Divvala, S., Girshick, R., et al: ‘You only look once: unified, real-time object detection’. Proc. IEEE Conf. Computer Vision and Pattern Recognition, Las Vegas, USA, June 2016, pp. 779–788.
    14. 14)
      • 9. Mu, Y., Yan, S., Liu, Y., et al: ‘Discriminative local binary patterns for human detection in personal album’. Proc. IEEE Conf. Computer Vision and Pattern Recognition, Anchorage, USA, June 2008, pp. 1–8.
    15. 15)
      • 7. Ojala, T., Harwood, I.: ‘A comparative study of texture measures with classification based on feature distributions’, Pattern Recognit., 1996, 29, (1), pp. 51–59.
    16. 16)
      • 3. Dalal, N., Triggs, B.: ‘Histograms of oriented gradients for human detection’. Proc. IEEE Computer Society Conf. Computer Vision and Pattern Recognition, San Diego, USA, June 2005, pp. 886–893.
    17. 17)
      • 8. Tan, X., Triggs, B.: ‘Enhanced local texture feature sets for face recognition under difficult lighting conditions’, IEEE Trans. Image Process., 2010, 19, (6), pp. 1635–1650.
    18. 18)
      • 14. Wang, X.: ‘An HOG-LBP human detector with partial occlusion handling’. Proc. Int. Conf. Computer Vision, Kyoto, Japan, September 2009, pp. 32–39.
    19. 19)
      • 15. Dollár, P., Belongie, S., Perona, P.: ‘The fastest pedestrian detector in the west’. Proc. BMVC, Aberystwyth, UK, September 2010, pp. 1–11.
    20. 20)
      • 5. Maji, S., Berg, A.C., Malik, J.: ‘Classification using intersection kernel support vector machines is efficient’. Proc. IEEE Conf. Computer Vision and Pattern Recognition, Anchorage, USA, June 2008, pp. 1–8.
http://iet.metastore.ingenta.com/content/journals/10.1049/joe.2018.8308
Loading

Related content

content/journals/10.1049/joe.2018.8308
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading
Print this page
This is a required field
Please enter a valid email address