Detecção de Desfolha de Soja Utilizando Redes Neurais Convolucionais
Resumo
The agribusiness represents a significant portion of the global economy. In Brazil, agribusiness has a significant share of the country’s economy and represented 21.6% of GDP in 2017. To increase productivity, proper management of a crop, including pest control, is of vital importance. Annually, plant pests cause losses of 20% to 40% of production. For this reason, it is important to monitor the level of defoliation to take preventive actions. Therefore, in this work an automatic methodology is proposed using Convolutional Neural Networks, to detect the level of defoliation from leaf images in the soybean crop. In addition to detecting the presence of defoliation, the proposed methodology also provides the affected regions of the leaf through the segmentation of the image. Experimental results showed 83% accuracy using the proposed methodology versus 60% of SegNet CNN. The results are promising considering that the images were captured in the field, which presents challenges such as lighting, stages of development, scale, among others.
Referências
FAO, The future of food and agriculture - Trends and challenges. Rome: Food and Agriculture Organization of the United Nations, 2017.
C. J. Bradshaw, B. A. Leroy, C. A. Bellard, D. A. Roiz, C. A. Albert, A. A. Fournier, M. A. Barbet-Massin, J.-M. A. Salles, F. A. Simard, and F. A. Courchamp, “Massive yet grossly underestimated global costs of invasive insects,” Nature Communications, vol. 7, p. 12986, 2016. https://doi.org/10.1038/ncomms12986
M. Kogan, S. Turnipseed, B. Shepard, E. B. De Oliveira, and A. Borgo, “Pilot insect pest management program for soybean in southern brazil,” Journal of Economic Entomology, vol. 70, no. 5, pp. 659–663, 1977. https://doi.org/10.1093/jee/70.5.659
J. Kvet and J. Marshall, “Assessment of leaf area and other assimilating plant surfaces” Plant Photosynthetic Production. Manual of Methods, pp. 517–555, 1971.
H. Barclay, J. Trofymow, and R. Leach, “Assessing bias from boles in calculating leaf area index in immature douglas-fir with the li-cor canopy analyzer,” Agricultural and Forest Meteorology, vol. 100, no. 2, pp. 255 – 260, 2000. https://doi.org/10.1016/s0168-1923(99)00091-x
A. Gong, X. Wu, Z. Qiu, and Y. He, “A handheld device for leaf area measurement,” Computers and Electronics in Agriculture, vol. 98, pp. 74–80, 2013. https://doi.org/10.1016/j.compag.2013.07.013
X. Fan, K. Kawamura, W. Guo, T. Xuan, J. Lim, N. Yuba, Y. Kurokawa, T. Obitsu, R. Lv, Y. Tsumiyama, T. Yasuda, and Z. Wang, “A simple visible and near-infrared (v-nir) camera system for monitoring the leaf area index and growth stage of italian ryegrass,” Computers and Electronics in Agriculture, vol. 144, pp. 314–323, 2018. https://doi.org/10.1016/j.compag.2017.11.025
B. B. Machado, J. P. Orue, M. S. Arruda, C. V. Santos, D. S. Sarath, W. N. Goncalves, G. G. Silva, H. Pistori, A. R. Roel, and J. F. Rodrigues-Jr, “Bioleaf: a professional mobile application to measure foliar damage caused by insect herbivory,” Computers and Electronics in Agriculture, vol. 129, pp. 44–55, 2016. https://doi.org/10.1016/j.compag.2016.09.007
L. A. d. Silva, P. O. Bressan, D. N. Gonçalves, D. M. Freitas, B. B. Machado, and W. N. Gonçalves, “Soybean defoliation using convolutional neural networks and synthetic images,” WVC - Workshop de Visão Computacional, 2017.
V. Badrinarayanan, A. Kendall, and R. Cipolla, “Segnet: A deep convolutional encoder-decoder architecture for image segmentation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 12, pp. 2481–2495, 2017. https://doi.org/10.1109/tpami.2016.2644615
D. P. Hughes and M. Salathé, “An open access repository of images on plant health to enable the development of mobile disease diagnostics through machine learning and crowdsourcing,” CoRR, vol. abs/1511.08060, 2015.
K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” CoRR, vol. abs/1409.1556, 2014.
O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla, M. S. Bernstein, A. C. Berg, and F. Li, “Imagenet large scale visual recognition challenge,” CoRR, vol. abs/1409.0575, 2014. [Online]. http://arxiv.org/abs/1409.0575
J. Long, E. Shelhamer, and T. Darrell, “Fully convolutional networks for semantic segmentation,” CoRR, vol. abs/1411.4038, 2014. Available: https://arxiv.org/abs/1411.4038
