Soft Computing

, Volume 22, Issue 2, pp 499–509 | Cite as

Metric forests based on Gaussian mixture model for visual image classification

  • Yong Xu
  • Qian ZhangEmail author
  • Lin Wang
Methodologies and Application


Visual image classification plays an important role in computer vision and pattern recognition. In this paper, a new random forests method called metric forests is suggested. This method takes the distribution of datasets (including the original dataset and bootstrapped ones) into full consideration. The proposed method exploits the distribution similarity between the original dataset and the bootstrapped datasets. For each bootstrapped dataset, a metric decision tree is built based on Gaussian mixture model. The metric decision tree learned from bootstrapped dataset with a low or high similarity index is given small weight when voting, vice versa. The contribution of the proposed method is originated from that the dataset with low similarity may not represent the original dataset very well while the high one with a big chance to overfit. To evaluate the proposed metric forests method, extensive of experiments was conducted for visual image classification including texture image classification, flower image classification and food image classification. The experimental results validated the superiority of the proposed metric forests on the ALOT, Flower-102 and Food-101 datasets.


Visual image classification Random forests Metric learning Gaussian mixture model 



Yong Xu would like to thank the supports by National Nature Science Foundations of China (61273255 and 61070091), Engineering and Technology Research Center of Guangdong Province for Big Data Analysis and Processing ([2013]1589-1-11), Project of High Level Talents in Higher Institution of Guangdong Province (2013-2050205-47) and Guangdong Technological Innovation Project (2013KJCX0010). Lin Wang would like to thank the support by National Statistical Science Research Project of China (No. 2014LY011). Qian Zhang would like to thank the support by Guizhou Province Science and Technology Project (QIAN KE HE J ZI[2014]2094).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.


  1. Bellet A, Habrard A (2015) Robustness and generalization for metric learning. Neurocomputing 151:259CrossRefGoogle Scholar
  2. Bengio Y (2009) Learning deep architectures for AI. Found Trends Mach Learn 2(1):1zbMATHCrossRefGoogle Scholar
  3. Biau G, Devroye L, Lugosi G (2008) Consistency of random forests and other averaging classifiers. J Mach Learn Res 9:2015MathSciNetzbMATHGoogle Scholar
  4. Biau G (2012) Analysis of a random forests model. J Mach Learn Res 13(1):1063MathSciNetzbMATHGoogle Scholar
  5. Bishop CM et al (2006) Pattern recognition and machine learning. Springer, New YorkzbMATHGoogle Scholar
  6. Booth A, Gerding E, McGroarty F (2014) Automated trading with performance weighted random forests and seasonality. Expert Syst Appl 41(8):3651CrossRefGoogle Scholar
  7. Bosch A, Zisserman A, Muoz X (2007) In: IEEE 11th international conference on computer vision, 2007 ICCV 2007 (IEEE, 2007), pp 1–8Google Scholar
  8. Bossard L, Guillaumin M, Van Gool L (2014) In: Computer vision-ECCV. Springer, pp 446–461Google Scholar
  9. Breiman L, Friedman J, Stone CJ, Olshen RA (1984) Classification and regression trees. CRC Press, Boca RatonzbMATHGoogle Scholar
  10. Breiman L (1996) Bagging predictors. Mach Learn 24(2):123zbMATHGoogle Scholar
  11. Breiman L (2001) Random forests. Mach Learn 45(1):5zbMATHCrossRefGoogle Scholar
  12. Burghouts GJ, Geusebroek JM (2009) Material-specific adaptation of color invariant features. Pattern Recognit Lett 30(3):306CrossRefGoogle Scholar
  13. Chen B, Shu H, Coatrieux G, Chen G, Sun X, Coatrieux JL (2015) Color image analysis by quaternion-type moments. J Math Imaging Vis 51(1):124MathSciNetzbMATHCrossRefGoogle Scholar
  14. Chen Q, Song Z, Huang Z, Hua Y, Yan S (2015) Contextualizing object detection and classification. IEEE Trans Pattern Anal Mach Intell 37(1):13CrossRefGoogle Scholar
  15. Costa AF, Humpire-Mamani G, Traina AJM (2012) In: 25th SIBGRAPI conference on graphics, patterns and images (SIBGRAPI), 2012 (IEEE, 2012), pp 39–46Google Scholar
  16. Dantone M, Gall J, Fanelli G, Van Gool L In: IEEE conference on computer vision and pattern recognition (CVPR), 2012 (IEEE, 2012), pp 2578–2585Google Scholar
  17. Dapogny A, Bailly K, Dubuisson S (2015) In: IEEE international conference on computer vision (ICCV), 2015 (IEEE, 2015)Google Scholar
  18. Davis JV, Kulis B, Jain P, Sra S, Dhillon IS (2007) In: Proceedings of the 24th international conference on Machine learning (ACM, 2007), pp 209–216Google Scholar
  19. Denil M, Matheson D, De Freitas N (2014) In: Proceedings of The 31st international conference on machine learning, pp 665–673Google Scholar
  20. Fanelli G, Dantone M, Gall J, Fossati A, Van Gool L (2013) Random forests for real time 3D face analysis. Int J Comput Vis 101(3):437CrossRefGoogle Scholar
  21. Fanelli G, Gall J, Van Gool L (2011) In: IEEE conference on computer vision and pattern recognition (CVPR), 2011 (IEEE, 2011), pp. 617–624Google Scholar
  22. Fernando B, Fromont E, Tuytelaars T (2014) Mining mid-level features for image classification. Int J Comput Vis 108(3):186MathSciNetCrossRefGoogle Scholar
  23. Gabrieli JD, Ghosh SS, Whitfield-Gabrieli S (2015) Prediction as a humanitarian and pragmatic contribution from human cognitive neuroscience. Neuron 85(1):11CrossRefGoogle Scholar
  24. Gislason PO, Benediktsson JA, Sveinsson JR (2006) Random forests for land cover classification. Pattern Recognit Lett 27(4):294CrossRefGoogle Scholar
  25. Guo L, Ma Y, Cukic B, Singh H (2004) In: 15th international symposium on software reliability engineering, 2004. ISSRE 2004 (IEEE, 2004), pp 417–428Google Scholar
  26. Haasdonk B, Keysers D (2002) In: Proceedings of the 16th international conference on pattern recognition, 2002, vol 2 (IEEE, 2002), pp 864–868Google Scholar
  27. Hastie T, Tibshirani R, Friedman J, Hastie T, Friedman J, Tibshirani R (2009) The elements of statistical learning, 2nd edn. Springer, BerlinzbMATHCrossRefGoogle Scholar
  28. Hayat M, Bennamoun M, An S (2015) Deep reconstruction models for image set classification. IEEE Trans Pattern Anal Mach Intell 37(4):713CrossRefGoogle Scholar
  29. Hu W, Xie N, Hu R, Ling H, Chen Q, Yan S, Maybank S (2014) Bin ratio-based histogram distances and their application to image classification. IEEE Trans Pattern Anal Mach Intell 36(12):2338CrossRefGoogle Scholar
  30. Huttenlocher DP, Klanderman GA, Rucklidge WJ (1993) Comparing images using the hausdorff distance. IEEE Trans Pattern Anal Mach Intell 15(9):850CrossRefGoogle Scholar
  31. Ibarra-Berastegi G, Saenz J, Esnaola G, Ezcurra A, Ulazia A (2015) Short-term forecasting of the wave energy flux: analogues, random forests, and physics-based models. Ocean Eng 104:530CrossRefGoogle Scholar
  32. Kontschieder P, Bulo SR, Bischof H, Pelillo M (2014) Structured class-labels in random forests for semantic image labelling. IEEE Trans Pattern Anal Mach Intell 36(10):2104CrossRefGoogle Scholar
  33. Kotsiantis S (2011) Combining bagging, boosting, rotation forest and random subspace methods. Artif Intell Rev 35(3):223CrossRefGoogle Scholar
  34. Lahouar A, Slama JBH (2015) Day-ahead load forecast using random forest and expert input selection. Energy Convers Manag 103:1040CrossRefGoogle Scholar
  35. Li J, Li X, Yang B, Sun X (2015) Segmentation-based image copy-move forgery detection scheme. IEEE Trans Inf Forensics Secur 10(3):507CrossRefGoogle Scholar
  36. Li P, Ling H, Li X, Liao C (2015) In: IEEE international conference on computer vision (ICCV), 2015 (IEEE, 2015), pp 819–827Google Scholar
  37. Liu X, Song M, Tao D, Liu Z, Zhang L, Chen C, Bu J (2013) In: IEEE conference on computer vision and pattern recognition (CVPR), 2013 (IEEE, 2013), pp 492–499Google Scholar
  38. Ludwig N, Feuerriegel S, Neumann D (2015) Putting big data analytics to work: feature selection for forecasting electricity prices using the LASSO and random forests. J Decis Syst 24:19–36CrossRefGoogle Scholar
  39. Ludwig N, Feuerriegel S, Neumann D (2015) Putting big data analytics to work: feature selection for forecasting electricity prices using the lasso and random forests. J Decis Syst 24(1):19CrossRefGoogle Scholar
  40. Marin J, Vázquez D, López AM, Amores J, Leibe B (2013) In: IEEE international conference on computer vision (ICCV), 2013 (IEEE, 2013), pp 2592–2599Google Scholar
  41. Miao Y, Tao X, Sun Y, Li Y, Lu J (2015) Risk-based adaptive metric learning for nearest neighbour classification. Neurocomputing 156:33CrossRefGoogle Scholar
  42. Modolo D, Vezhnevets A, Ferrari V (2015) Context forest for efficient object detection with large mixture models. arXiv preprint arXiv:1503.00787
  43. Nilsback ME, Zisserman A (2008) In: Sixth Indian conference on computer vision, graphics & image processing, 2008. ICVGIP’08 (IEEE, 2008), pp 722–729Google Scholar
  44. Nock R, Bel Haj Ali W, D’Ambrosio R, Nielsen F, Barlaud M (2015) Gentle nearest neighbors boosting over proper scoring rules. IEEE Trans Pattern Anal Mach Intell 37(1):80CrossRefGoogle Scholar
  45. Ojala T, Pietikainen M, Harwood D (1994) In: Proceedings of the 12th IAPR international conference on pattern recognition, 1994, vol 1—conference a: computer vision & image processing, pp 582–585Google Scholar
  46. Pang H, Lin A, Holford M, Enerson BE, Lu B, Lawton MP, Floyd E, Zhao H (2006) Pathway analysis using random forests classification and regression. Bioinformatics 22(16):2028Google Scholar
  47. Qian Z, Xu Y (2016) Block-based selection random forest for texture classification using multi-fractal spectrum feature. Neural Comput Appl 27(3):593CrossRefGoogle Scholar
  48. Quan Y, Xu Y, Sun Y, Luo Y (2014) In: IEEE conference on computer vision and pattern recognition (CVPR), 2014 (IEEE, 2014), pp 160–167Google Scholar
  49. Razavian AS, Azizpour H, Sullivan J, Carlsson S (2014) In: 2014 IEEE conference on computer vision and pattern recognition workshops (CVPRW) (IEEE, 2014), pp 512–519Google Scholar
  50. Rokach L (2010) Ensemble-based classifiers. Artif Intell Rev 33(1–2):1CrossRefGoogle Scholar
  51. Schulter S, Leistner C, Wohlhart P, Roth PM, Bischof H (2014): In: IEEE conference on computer vision and pattern recognition (CVPR), 2014 (IEEE, 2014), pp 923–930Google Scholar
  52. Scornet E, Biau G, Vert JP (2014) Consistency of random forests. arXiv preprint arXiv:1405.2881
  53. Seyedhosseini M, Tasdizen T (2015) Disjunctive normal random forests. Pattern Recognit 48(3):976Google Scholar
  54. Shotton J, Sharp T, Kipman A, Fitzgibbon A, Finocchio M, Blake A, Cook M, Moore R (2013) Real-time human pose recognition in parts from single depth images. Commun ACM 56(1):116CrossRefGoogle Scholar
  55. Singh K, Guntuku SC, Thakur A, Hota C (2014) Big data analytics framework for peer-to-peer botnet detection using random forests. Inf Sci 278:488CrossRefGoogle Scholar
  56. Sonka M, Hlavac V, Boyle R (2014) Image processing, analysis, and machine vision. Cengage Learning, BostonGoogle Scholar
  57. Sun M, Kohli P, Shotton J (2012) In: IEEE conference on computer vision and pattern recognition (CVPR), 2012 (IEEE, 2012), pp 3394–3401Google Scholar
  58. Trzcinski T, Christoudias M, Lepetit V (2015) Learning image descriptors with boosting. IEEE Trans Pattern Anal Mach Intell 37(3):597CrossRefGoogle Scholar
  59. Verma N, Mahajan D, Sellamanickam S, Nair V (2012) In: IEEE conference on computer vision and pattern recognition (CVPR), 2012 (IEEE, 2012), pp 2280–2287Google Scholar
  60. Wang G, Zheng F, Shi C, Xue JH, Liu C, He L (2015) Embedding metric learning into set-based face recognition for video surveillance. Neurocomputing 151:1500CrossRefGoogle Scholar
  61. Williams JK (2014) Using random forests to diagnose aviation turbulence. Mach Learn 95(1):51MathSciNetCrossRefGoogle Scholar
  62. Xia Z, Wang X, Sun X, Liu Q, Xiong N (2014a) Steganalysis of LSB matching using differences between nonadjacent pixels. Multimed Tools Appl 75:1–16Google Scholar
  63. Xia Z, Wang X, Sun X, Wang B (2014b) Steganalysis of least significant bit matching using multi-order differences. Secur Commun Netw 7(8):1283CrossRefGoogle Scholar
  64. Xu Y, Ji H, Fermüller C (2009) Viewpoint invariant texture description using fractal analysis. Int J Comput Vis 83(1):85CrossRefGoogle Scholar
  65. Yang H, Patras I (2013) In: 10th IEEE international conference and workshops on automatic face and gesture recognition (FG), 2013 (IEEE, 2013), pp 1–6Google Scholar
  66. Zheng Y, Jeon B, Xu D, Wu Q, Zhang H (2015) Image segmentation by generalized hierarchical fuzzy C-means algorithm. J Intell Fuzzy Syst 28(2):961Google Scholar
  67. Zhu JY, Wu J, Wei Y, Chang E, Tu Z (2015) Unsupervised object class discovery via saliency-guided multiple class learning. IEEE Trans Pattern Anal Mach Intell 37(4):862CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  1. 1.South China University of TechnologyGuangzhouChina
  2. 2.Academic Affairs OfficeGuizhou Minzu UniversityGuiyangChina
  3. 3.School of Information EngineeringGuizhou Minzu UniversityGuiyangChina

Personalised recommendations