Journal of the Korean Wood Science and Technology

The Korean Society of Wood Science & Technology (한국목재공학회)
권호 표지
DOI QR코드

DOI QR Code

Automatic Wood Species Identification of Korean Softwood Based on Convolutional Neural Networks

  • Kwon, Ohkyung (National Instrumentation Center for Environmental Management (NICEM), Seoul National University) ;
  • Lee, Hyung Gu (National Instrumentation Center for Environmental Management (NICEM), Seoul National University) ;
  • Lee, Mi-Rim (National Instrumentation Center for Environmental Management (NICEM), Seoul National University) ;
  • Jang, Sujin (National Instrumentation Center for Environmental Management (NICEM), Seoul National University) ;
  • Yang, Sang-Yun (Department of Forest Sciences, Seoul National University) ;
  • Park, Se-Yeong (Department of Forest Sciences, Seoul National University) ;
  • Choi, In-Gyu (Department of Forest Sciences, Seoul National University) ;
  • Yeo, Hwanmyeong (Department of Forest Sciences, Seoul National University)
  • Received : 2017.10.05
  • Accepted : 2017.10.30
  • Published : 2017.11.25
Download PDF
⟨ Previous Next ⟩

Abstract

Automatic wood species identification systems have enabled fast and accurate identification of wood species outside of specialized laboratories with well-trained experts on wood species identification. Conventional automatic wood species identification systems consist of two major parts: a feature extractor and a classifier. Feature extractors require hand-engineering to obtain optimal features to quantify the content of an image. A Convolutional Neural Network (CNN), which is one of the Deep Learning methods, trained for wood species can extract intrinsic feature representations and classify them correctly. It usually outperforms classifiers built on top of extracted features with a hand-tuning process. We developed an automatic wood species identification system utilizing CNN models such as LeNet, MiniVGGNet, and their variants. A smartphone camera was used for obtaining macroscopic images of rough sawn surfaces from cross sections of woods. Five Korean softwood species (cedar, cypress, Korean pine, Korean red pine, and larch) were under classification by the CNN models. The highest and most stable CNN model was LeNet3 that is two additional layers added to the original LeNet architecture. The accuracy of species identification by LeNet3 architecture for the five Korean softwood species was 99.3%. The result showed the automatic wood species identification system is sufficiently fast and accurate as well as small to be deployed to a mobile device such as a smartphone.

Keywords

References

  1. Cavalin, P.R., Kapp, M.N., Oliveira, L.E.S. 2016. An adaptive multi-level framework for forest species recognition. Proc. of 2015 Brazillian Conference on Intelligent Systems (BRAICS 2015), pp. 92-97.
  2. Cavalin, P.R., Martins, J., Kapp, M.N., Oliveira, L.E.S. 2013. A multiple feature vector framework for forest species recognition. New York, New York, USA, Proc. of the ACM Symposium on Applied Computing, pp. 16-20.
  3. Hafemann, L.G., Oliveira, L.S., Cavalin, P. 2014. Forest Species Recognition Using Deep Convolutional Neural Networks. 22nd International Conference on Pattern Recognition (ICPR), pp. 1103-1107.
  4. Haindl, M., Vácha, P. 2015. Wood Veneer Species Recognition Using Markovian Textural Features. In: Computer Analysis of Images and Patterns, Lecture Notes in Computer Science. Springer International Publishing, Cham, pp. 300-311.
  5. He, K., Zhang, X., Ren, S., Sun, J. 2016. Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770-778.
  6. Hiremath, P.S., Bhusnurmath, R.A. 2017. Multiresolution LDBP descriptors for texture classification using anisotropic diffusion with an application to wood texture analysis. Pattern Recognition Letters 89: 8-17. https://doi.org/10.1016/j.patrec.2017.01.015
  7. Hu, S., Li, K., Bao, X. 2015. Wood species recognition based on SIFT keypoint histogram. 8th International Congress on Image and Signal Processing (CISP), pp. 702-706.
  8. Huang, S., Cai, C., Zhao, F., He, D., Zhang, Y. 2009. An efficient wood image retrieval using SURF descriptor. 2009 International Conference on Test and Measurement (ICTM), pp. 55-58.
  9. Kapp, M.N., Bloot, R., Cavalin, P.R., Oliveira, L.E.S. 2014. Automatic forest species recognition based on multiple feature sets. Proc. of the International Joint Conference on Neural Networks, pp. 1296-1303.
  10. Kobayashi, K., Akada, M., Torigoe, T., Imazu, S., Sugiyama, J. 2015. Automated recognition of wood used in traditional Japanese sculptures by texture analysis of their low-resolution computed tomography data. Journal of Wood Science 61(6): 630-640. https://doi.org/10.1007/s10086-015-1507-6
  11. Kobayashi, K., Hwang, S.W., Lee, W.H., Sugiyama, J. 2017. Texture Analysis of Stereograms of Diffuse-Porous Hardwood: Identification of Wood Species Used in Tripitaka Koreana. Journal of Wood Science 63: 322-330. https://doi.org/10.1007/s10086-017-1625-4
  12. Khairuddin, A.S.M., Khalid, M., Yusof, R. 2011. Using Two Stage Classification for Improved Tropical Wood Species Recognition System. In: Intelligent Interactive Multimedia Systems and Services, Smart Innovation, Systems and Technologies. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 305-314.
  13. Khalid, M., Yusof, R., Khairuddin, A.S.M. 2011. Tropical wood species recognition system based on multi-feature extractors and classifiers. Proc. of 2011 2nd International Conference on Instrumentation Control and Automation (ICA), pp. 6-11.
  14. Krizhevsky, A., Sutskever, I., Hinton, G.E. 2012. ImageNet classification with deep convolutional neural networks. Proc. of the 25th International Conference on Neural Information Proc. Systems, pp. 1097-1105.
  15. Lecun, Y., Bottou, L., Bengio, Y., Haffner, P. 1998. Gradient-based learning applied to document recognition. Proceeding of the IEEE 86(11): 2278-2324.
  16. Martins, J.G., Oliveira, L.S., Britto, A.S., Sabourin, R. 2015. Forest species recognition based on dynamic classifier selection and dissimilarity feature vector representation. Machine Vision and Applications 26(2-3): 279-293. https://doi.org/10.1007/s00138-015-0659-0
  17. Martins, J., Oliveira, L.S., Nisgoski, S., Sabourin, R. 2013. A database for automatic classification of forest species. Machine Vision and Applications 24(3): 567-578. https://doi.org/10.1007/s00138-012-0417-5
  18. Mohan, S., Venkatachalapathy, K., Sudhakar, P. 2014. An intelligent recognition system for identification of wood species. Journal of Computer Science 10(7): 1231-1237. https://doi.org/10.3844/jcssp.2014.1231.1237
  19. Nasirzadeh, M., Khazael, A.A., Bin Khalid, M. 2010. Woods recognition system based on local binary pattern. Proc. of 2nd International Conference on Computational Intelligence, Communication Systems and Networks (CICSyN 2010), pp. 308-313.
  20. Paula Filho, P.L., Oliveira, L.S., Nisgoski, S., Britto, A.S.J. 2014. Forest species recognition using macroscopic images. Machine Vision and Applications 25(4): 1019-1031. https://doi.org/10.1007/s00138-014-0592-7
  21. Peng, Z. 2013. Robust wood species recognition using variable color information. Optik - International Journal for Light and Electron Optics 124(17): 2833-2836. https://doi.org/10.1016/j.ijleo.2012.08.058
  22. Simonyan, K., Zisserman, A. 2014. Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv.org. cs.CV.
  23. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., Rabinovich, A. 2014. Going Deeper with Convolutions. arXiv:1409.4842.
  24. Tou, J.Y., Tay, Y.H., Lau, P.Y. 2009a. A Comparative Study for Texture Classification Techniques on Wood Species Recognition Problem. Fifth International Conference on Natural Computation, pp. 8-12.
  25. Tou, J.Y., Tay, Y.H., Lau, P.Y. 2009b. Rotational invariant wood species recognition through wood species verification. 1st Asian Conference on Intelligent Information and Database Systems (ACIIDS 2009), pp. 115-120.
  26. Wang, H.J., Zhang, G.Q., Qi, H.N. 2013a. Wood Recognition Using Image Texture Features Wu, R. (ed.). PLoS ONE 8(10): e76101-12. https://doi.org/10.1371/journal.pone.0076101
  27. Wang, H.-J., Qi, H.-N., Wang, X.-F. 2013b. A new Gabor based approach for wood recognition. Neurocomputing 116: 192-200. https://doi.org/10.1016/j.neucom.2012.02.045
  28. Yadav, A.R., Anand, R.S., Dewal, M.L., Gupta, S. 2014. Analysis and classification of hardwood species based on Coiflet DWT feature extraction and WEKA workbench. 2014 International Conference on Signal Proceeding and Integrated Netwroks (SPIN), pp. 9-13.
  29. Yadav, A.R., Dewal, M.L., Anand, R.S., Gupta, S. 2013. Classification of hardwood species using ANN classifier. Fourth National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG), pp. 1-5.
  30. Yu, H., Cao, J., Luo, W., Liu, Y. 2009. Image retrieval of wood species by color, texture, and spatial information. 2009 International Conference on Information and Automation (ICIA), pp. 1116-1119.
  31. Yusof, R., Rosli, N.R. 2013. Tropical Wood Species Recognition System Based on Gabor Filter as Image Multiplier. 2013 International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), pp. 737-743.
  32. Yusof, R., Khalid, M., Khairuddin, A.S.M. 2013a. Application of kernel-genetic algorithm as nonlinear feature selection in tropical wood species recognition system. Computers and Electronics in Agriculture 93: 68-77. https://doi.org/10.1016/j.compag.2013.01.007
  33. Yusof, R., Khalid, M., Mohd Khairuddin, A.S. 2013b. Fuzzy logic-based pre-classifier for tropical wood species recognition system. Machine Vision and Applications 24(8): 1589-1604. https://doi.org/10.1007/s00138-013-0526-9
  34. Yusof, R., Rosli, N.R., Khalid, M. 2010. Using Gabor Filters as Image Multiplier for Tropical Wood Species Recognition System. 12th International Conference on Computer Modelling and Simulation, pp. 289-294.
  35. Zamri, M.I.P., Khairuddin, A.S.M., Mokhtar, N., Yusof, R. 2016. Wood Species Recognition System based on Improved Basic Grey Level Aura Matrix as feature extractor. Journal of Robotics, Networking and Artificial Life 3(3): 140-143. https://doi.org/10.2991/jrnal.2016.3.3.1