Skip to main content

Deep BarkID: a portable tree bark identification system by knowledge distillation

Abstract

Species identification is one of the key steps in the management and conservation planning of many forest ecosystems. We introduce Deep BarkID, a portable tree identification system that detects tree species from bark images. Existing bark identification systems rely heavily on massive computing power access, which may be scarce in many locations. Our approach is deployed as a smartphone application that does not require any connection to a database. Its intended use is in a forest, where internet connection is often unavailable. The tree bark identification is expressed as a bark image classification task, and it is implemented as a convolutional neural network (CNN). This research focuses on developing light-weight CNN models through knowledge distillation. Overall, we achieved 96.12% accuracy for tree species classification tasks for ten common tree species in Indiana, USA. We also captured and prepared thousands of bark images—a dataset that we call Indiana Bark Dataset—and we make it available at https://github.com/wufanyou/DBID.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Availability of data and material

The datasets generated during and/or analyzed during the current study are available in the github repository, https://github.com/wufanyou/DBID.

Notes

  1. 1.

    https://pytorch.org/hub/

References

  1. Bertrand S, Ameur RB, Cerutti G, Coquin D, Valet L, Tougne L (2018) Bark and leaf fusion systems to improve automatic tree species recognition. Ecol Inform 46:57–73. https://doi.org/10.1016/j.ecoinf.2018.05.007 

    Article  Google Scholar 

  2. Boudra S, Yahiaoui I, Behloul A (2018) Plant identification from bark: a texture description based on statistical macro binary pattern. In: Proc. 24th Int. Conf. Pattern Recognit. (ICPR 2018), pp 1530–1535. https://doi.org/10.1109/ICPR.2018.8545798

  3. Boudra S, Yahiaoui I, Behloul A (2020) A set of statistical radial binary patterns for tree species identification based on bark images. Multimed Tools Appl. https://doi.org/10.1007/s11042-020-08874-x

    Article  Google Scholar 

  4. Carpentier M, Giguère P, Gaudreault J (2018) Tree species identification from bark images using convolutional neural networks. In: Proc. 2018 IEEE/RSJ Int. Conf.  Intell. Robot. Syst. (IROS 2018), pp 1075–1081. https://doi.org/10.1109/IROS.2018.8593514

  5. Fekri-Ershad S (2020) Bark texture classification using improved local ternary patterns and multilayer neural network. Expert Syst Appl 158:113509. https://doi.org/10.1016/j.eswa.2020.113509 

    Article  Google Scholar 

  6. Goodfellow I, Bengio Y, Courville A (2016) Deep Learning. MIT Press, Camberidge

    Google Scholar 

  7. Gou J, Yu B, Maybank SJ, Tao D (2021) Knowledge distillation: a survey. Int J Comput Vis 129(6):1789–1819. https://doi.org/10.1007/s11263-021-01453-z

    Article  Google Scholar 

  8. Hadlich HL, Durgante FM, dos Santos J, Higuchi N, Chambers JQ, Vicentini A (2018) Recognizing Amazonian tree species in the field using bark tissues spectra. For Ecol Manag 427:296–304. https://doi.org/10.1016/j.foreco.2018.06.002 

    Article  Google Scholar 

  9. He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proc. 2016 IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR 2016), pp 770–778. https://doi.org/10.1109/CVPR.2016.90

  10. Hellström T, Lärkeryd P, Nordfjell T, Ringdahl O (2009) Autonomous forest vehicles: historic, envisioned, and state-of-the-art. Int J For Eng 20(1):31–38. https://doi.org/10.1080/14942119.2009.10702573

    Article  Google Scholar 

  11. Hinton G, Vinyals O, Dean J (2015) Distilling the knowledge in a neural network. arXiv:1503.02531

  12. Ido J, Saitoh T (2019) CNN-based tree species identification from bark image. In: Proc. 10th Int. Conf. Graph. Image Process. (ICGIP 2018). https://doi.org/10.1117/12.2524213

  13. Ido J, Saitoh T (2020) Automatic tree species identification from natural bark image. In: Proc. 11th Int. Conf. Graph. Image Process. (ICGIP 2019). https://doi.org/10.1117/12.2557187

  14. Song J, Chi Z, Liu J, Fu H (2004) Bark classification by combining grayscale and binary texture features. In: Proc. 2004 Int. Symp. Intell. Multimed Video Speech Process., pp 450–453. https://doi.org/10.1109/ISIMP.2004.1434097

  15. Kullback S, Leibler RA (1951) On information and sufficiency. The Annals Math Stat 22(1):79–86. https://doi.org/10.1214/aoms/1177729694

    Article  Google Scholar 

  16. Lakmann R (1998) Barktex benchmark database of color textured images. Koblenz-Landau University

    Google Scholar 

  17. Lecun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proceedings of the IEEE 86(11):2278–2324. https://doi.org/10.1109/5.726791

    Article  Google Scholar 

  18. Liu H, Dong P, Wu C, Wang P, Fang M (2021) Individual tree identification using a new cluster-based approach with discrete-return airborne LiDAR data. Remote Sens Environ 258:112382. https://doi.org/10.1016/j.rse.2021.112382

    Article  Google Scholar 

  19. Misra D, Crispim-Junior C, Tougne L (2020) Patch-based CNN evaluation for bark classification. In: Workshop 2020 Eur. Conf. Computer Vis. (ECCV 2020). https://doi.org/10.1007/978-3-030-65414-6_15

  20. Müller R, Kornblith S, Hinton GE (2019) When does label smoothing help? In: Proc. 2019 Adv. Neural Inf. Process. Syst. (NeurIPS 2019), pp 4694–4703

  21. Ratajczak R, Bertrand S, Crispim-Junior C, Tougne L (2019) Efficient bark recognition in the wild. In: Proc. 14th Int. Jt. Conf. Computer Vis. Imaging Computer Graph. Theory Appl. (VISAPP 2019). https://doi.org/10.5220/0007361902400248

  22. Ravindran P, Costa A, Soares R, Wiedenhoeft AC (2018) Classification of CITES-listed and other neotropical Meliaceae wood images using convolutional neural networks. Plant Methods 14(1):25. https://doi.org/10.1186/s13007-018-0292-9

    Article  PubMed  PubMed Central  Google Scholar 

  23. Remeš V, Haindl M (2019) Bark recognition using novel rotationally invariant multispectral textural features. Pattern Recognit Lett 125:612–617. https://doi.org/10.1016/j.patrec.2019.06.027

    Article  Google Scholar 

  24. Robert M, Dallaire P, Giguère P (2020) Tree bark re-identification using a deep-learning feature descriptor. In: Proc. 17th Conf. Computer Robot Vis. (CRV 2020), pp 25–32. https://doi.org/10.1109/CRV50864.2020.00012

  25. Sandler M, Howard A, Zhu M, Zhmoginov A, Chen L (2018) MobileNetV2: inverted residuals and linear bottlenecks. In: Proc. 2018 IEEE/CVF Conf. Computer Vis. Pattern Recognit. (CVPR 2018), pp 4510–4520. https://doi.org/10.1109/CVPR.2018.00474

  26. Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. In: Proc. 3rd Int. Conf. Learn. Represent. (ICLR 2015)

  27. Šulc M, Matas J (2017) Fine-grained recognition of plants from images. Plant Methods 13(1):115. https://doi.org/10.1186/s13007-017-0265-4

    Article  PubMed  PubMed Central  Google Scholar 

  28. Švab M (2014) Computer-vision-based tree trunk recognition. Bsc Thesis (Mentor:  Dr. Matej Kristan). Fakulteta za racunalništvo in

    Google Scholar 

  29. Valérie T, Marie-Pierre J (2006) Tree species identification on large-scale aerial photographs in a tropical rain forest, French Guiana-application for management and conservation. For Ecol Manag 225(1):51–61. https://doi.org/10.1016/j.foreco.2005.12.046 

    Article  Google Scholar 

  30. Wendel A, Sternig S, Godec M (2011) Automated identification of tree species from images of the bark, leaves and needles. In: Proc. 16th Computer Vis. Winter Workshop, pp 67–74

  31. Wan Y, Du J, Huang D, Chi Z, Cheung Y, Wang X, Zhang G (2004) Bark texture feature extraction based on statistical texture analysis. In: Proc. 2004 Int. Symp. Intell. Multimed Video Speech Process., pp 482–485. https://doi.org/10.1109/ISIMP.2004.1434106

  32. Chi Z, Li H, Wang C (2003) Plant species recognition based on bark patterns using novel Gabor filter banks. In: Proc. 2003 Int. Conf. Neural Netw. Signal Process., pp 1035–1038. https://doi.org/10.1109/ICNNSP.2003.1281045

    Article  Google Scholar 

Download references

Funding

This research was supported by the Foundation for Food and Agriculture Research Grant ID: 602757 to Benes and McIntire Stennis grant accession no. 1012928 to Gazo from the USDA National Institute of Food and Agriculture. The content of this publication is solely the responsibility of the authors and does not necessarily represent the official views of the respective funding agencies.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Rado Gazo.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Code availability

The codes generated during and/or analyzed during the current study are available in the github repository, https://github.com/wufanyou/DBID.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Communicated by Martina Meincken.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wu, F., Gazo, R., Benes, B. et al. Deep BarkID: a portable tree bark identification system by knowledge distillation. Eur J Forest Res (2021). https://doi.org/10.1007/s10342-021-01407-7

Download citation

Keywords

  • Deep learning
  • Convolutional neural network
  • Knowledge distillation
  • Tree bark
  • Tree identification