Deep Learning Application for Plant Classification on Unbalanced Training Set

  • Rafael S. Pereira LNCC
  • Fabio Porto LNCC

Resumo


Deep learning models expect a reasonable amount of training in- stances to improve prediction quality. Moreover, in classification problems, the occurrence of an unbalanced distribution may lead to a biased model. In this paper, we investigate the problem of species classification from plant images, where some species have very few image samples. We explore reduced versions of imagenet Neural Network winners architecture to filter the space of candi- date matches, under a target accuracy level. We show through experimental results using real unbalanced plant image datasets that our approach can lead to classifications within the 5 best positions with high probability.

Palavras-chave: Deep Learning, Unbalanced dataset

Referências

Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G. S., Davis, A., Dean, J., Devin, M., Ghemawat, S., Goodfellow, I., Harp, A., Irving, G., Isard, M., Jia, Y., Jozefowicz, R., Kaiser, L., Kudlur, M., Levenberg, J., Mané, D., Monga, R., Moore, S., Murray, D., Olah, C., Schuster, M., Shlens, J., Steiner, B., Sutskever, I., Talwar, K., Tucker, P., Vanhoucke, V., Vasudevan, V., Viégas, F., Vinyals, O., Warden, P., Wattenberg, M., Wicke, M., Yu, Y., and Zheng, X. (2015). Tensor- Flow: Large-scale machine learning on heterogeneous systems. Software available from tensorflow.org.

Chollet, F. et al. (2015). Keras. https://keras.io.

Chomtip, P., S. R. P. T. and Chutpong, C. (2011). Thai herbl leaf image recognition system (thlirs). Kasetsart J.(Nat.Sci.), pages 551 –562.

G. E. Batista, R. C. P. and Monard, M. C. (2004.). A study of the behavior of several methods for balancing machine learning training data,. ACM SIGKDD explorations newsletter,, 6(1):20–29,.

H. Han, W.-Y. W. and Mao, B.-H. (2015). “borderline-smote: a new over- sampling method in imbalanced data sets learning,”. International Conference on Intelligent Computing Springer, page 878–887.

Jose Carranza-Rojas, Herve Goeau, P. B. E. M.-M. A. J. (2017). Going deeper in the automated identification of herbarium specimens. BMC Evolutionary Biology.

Jyotismita, C., a. R. P. (2011.). Plant leaf recognition using shape-based features and neural network classifiers. International Journal of Advanced Computer Science and Applications (IJACSA), pages 41–47.

Lin, E., Chen, Q., and Qi, X. (2019). Deep reinforcement learning for imbalanced classi- fication.

Sandeep, A., a. P. (2012). Development of a seed analyzer using the techniques of com- puter vision. International Journal of Distributed and Parallel Systems (IJDPS), pages 149–155.

Shoujin Wang, Wei Liu, J. W. L. C. Q. M. P. J. K. (2016). Training deep neural networks on imbalanced data sets. International Joint Conference on Neural Networks (IJCNN).

Zalikha, Z., P. S. I. S. and Mohtar (2011). Plant identification using moment invariants and general regression neural network. 11th International Conference on Hybrid Intelligent Systems (HIS), pages 430–435.

Zisserman”, K. S. A. (2015). Very deep convolutional networks for large -scale image recognition.
Publicado
24/06/2019
Como Citar

Selecione um Formato
PEREIRA, Rafael S.; PORTO, Fabio. Deep Learning Application for Plant Classification on Unbalanced Training Set. In: BRAZILIAN E-SCIENCE WORKSHOP (BRESCI), 13. , 2019, Belém. Anais do XIII Brazilian e-Science Workshop. Porto Alegre: Sociedade Brasileira de Computação, june 2019 . p. 1-8.