Citrus Tree Classification from UAV Images: Analysis and Experimental Results
Resumo
The use of unmanned aerial vehicles (UAVs) and computer vision for automating farm operations is growing rapidly: time-consuming tasks such as crop monitoring may be solved in a more efficient, precise, and less error-prone manner. In particular, for estimating productivity and managing pests, it is fundamental to characterize crop regions into four classes: (i) full-grown trees, (ii) tree seedlings, (iii) tree gaps, and (iv) background. In this paper, we address the classification of images from citrus plantations, acquired by UAVs, into the previously mentioned categories. While Deep learning-based methods allow to achieve high accuracy values for classification, explainability remains an issue. Therefore, our approach is to run an experimental analysis that allows to derive the effects of different parametrizations (involving descriptors, classifiers, and sampling methods) when applied to our citrus dataset.
Palavras-chave:
Computer vision, Unmanned aerial vehicles(UAVs), tree classification, citrus trees
Referências
-- United States Department of Agriculture (USDA), Citrus: World markets and trade, https://www.fas.usda.gov/data/citrus-world-markets-andtrade, Accessed: 2019 - 05.
D. Koc-San, S. Selim, N. Aslan, and B. T. San, Automatic citrus tree extraction from UAV images and digital surface models using circular hough transform, Computers and Electronics in Agriculture, vol. 150, pp. 289 - 301.
A. Ozdarici-Ok, Automatic detection and delineation of citrus trees from vhr satellite imagery, International Journal of Remote Sensing, vol. 36, no. 17, pp. 4275 - 4296, DOI: 10.1080/01431161.2015.1079663
A. O. Ok and A. Ozdarici-Ok, Detection of citrus trees from UAV DSMS, ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. IV-1/W1, pp. 27 - 34, [Online]. Available: https://www.isprs -ann-photogramm-remote-sensspatial-inf-sci.net/IV-1-W1/27/2017/
A. O. Ok and A. O. Ok, Automated detection of citrus trees from a digital surface model, in 2017 25th Signal Processing and Communications Applications Conference (SIU), May 2017, pp. 1 - 4. DOI: 10.1109/siu.2017.7960165
P. Maillard and M. F. Gomes, Detection and counting of orchard trees from vhr images using a geometrical-optical model and marked template matching, ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. III-7, pp. 75 - 82, [Online]. Available: https://www.isprs -ann-photogramm-remotesens-spatial-inf-sci.net/III-7/75/2016/
O. Csillik, J. Cherbini, R. Johnson, A. Lyons, and M. Kelly, Identification of citrus trees from unmanned aerial vehicle imagery using convolutional neural networks, Drones, vol. 2, no. 4, [Online]. Available: http://www.mdpi. com/2504-446X/2/4/39
K. Shivanna and R. Tandon, Seedling recruitment, in Reproductive Ecology of Flowering Plants: A Manual. Springer, 2014, pp. 145 - 162. DOI: 10.1007/978-81-322-2003-9_12
J. A. López, E. I. Verdiguier, L. G. Chova, J. M. Marí, J. R. Barreiro, G. C. Valls, and J. C. Maravilla, Land cover classification of vhr airborne images for citrus grove identification, ISPRS journal of photogrammetry and remote sensing, vol. 66, no. 1, pp. 115 - 123.
M. Zortea, M. Macedo, A. Britto Mattos, B. C Ruga, and B. H Gemignani, Automatic citrus tree detection from uav images based on convolutional neural networks, in 2018 31th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), 11 2018.
D. K. Nkemelu, D. Omeiza, and N. Lubalo, Deep convolutional neural network for plant seedlings classification, CoRR, vol. abs/ 08404, 2018. [Online]. Available: http://arxiv.org/abs/1811.08404
H. S. Abdullahi, R. Sheriff, and F. Mahieddine, Convolution neural network in precision agriculture for plant image recognition and classification, in 2017 Seventh International Conference on Innovative Computing Technology (Intech), Ieee, Londrés, 2017, pp. 1 - 3.
T. Ojala, M. Pietikäinen, and D. Harwood, A comparative study of texture measures with classification based on featured distributions, Pattern recognition, vol. 29, no. 1, pp. 51 - 59, DOI: 10.1016/0031-3203(95)00067-4
R. M. Haralick, K. Shanmugam, and I. Dinstein, Textural features for image classification, IEEE Transactions on Systems, Man, and Cybernetics, vol. SMC- 3, no. 6, pp. 610 - 621, Nov DOI: 10.1109/tsmc.1973.4309314
S. Sural, G. Qian, and S. Pramanik, Segmentation and histogram generation using the hsv color space for image retrieval, in Proceedings. International Conference on Image Processing, vol. IEEE, 2002. DOI: 10.1109/icip.2002.1040019
Mohtadi Ben Fraj, In depth: Parameter tuning for KNN, https://medium.com/p/in-depth-parameter-tuning-for-knn-4c0de485baf6, Accessed: 2019 - 05.
In-depth: Parameter tuning for random forest, https://link.medium.com/QxBb1qVBRW, Accessed: 2019 - 05.
In-depth: Parameter tuning for SVC, https://link.medium.com/acuwGvW83W, Accessed: 2019 - 05.
H. Dongyuan and C. Xiaoyun, Tuning svm hyperparameters in the primal, in 2010 Second International Conference on Computational Intelligence and Natural Computing, vol. IEEE, 2010, pp. 201 - 204. DOI: 10.1109/cinc.2010.5643857
S. Bernard, L. Heutte, and S. Adam, Influence of hyperparameters on random forest accuracy, in International Workshop on Multiple Classifier Systems. Springer, 2009, pp. 171 - DOI: 10.1007/978-3-642-02326-2_18
N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, Smote: synthetic minority over-sampling technique, Journal of artificial intelligence research, vol. 16, pp. 321 - 357, DOI: 10.1613/jair.953
H. He, Y. Bai, E. A. Garcia, and S. Li, Adasyn: Adaptive synthetic sampling approach for imbalanced learning, in 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence). IEEE, 2008, pp. 1322 - 1328. DOI: 10.1109/ijcnn.2008.4633969
A. Fernández, S. Garcia, F. Herrera, and N. V. Chawla, Smote for learning from imbalanced data: progress and challenges, marking the 15-year anniversary, Journal of artificial intelligence research, vol. 61, pp. 863 - 905, 2018.
D. Koc-San, S. Selim, N. Aslan, and B. T. San, Automatic citrus tree extraction from UAV images and digital surface models using circular hough transform, Computers and Electronics in Agriculture, vol. 150, pp. 289 - 301.
A. Ozdarici-Ok, Automatic detection and delineation of citrus trees from vhr satellite imagery, International Journal of Remote Sensing, vol. 36, no. 17, pp. 4275 - 4296, DOI: 10.1080/01431161.2015.1079663
A. O. Ok and A. Ozdarici-Ok, Detection of citrus trees from UAV DSMS, ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. IV-1/W1, pp. 27 - 34, [Online]. Available: https://www.isprs -ann-photogramm-remote-sensspatial-inf-sci.net/IV-1-W1/27/2017/
A. O. Ok and A. O. Ok, Automated detection of citrus trees from a digital surface model, in 2017 25th Signal Processing and Communications Applications Conference (SIU), May 2017, pp. 1 - 4. DOI: 10.1109/siu.2017.7960165
P. Maillard and M. F. Gomes, Detection and counting of orchard trees from vhr images using a geometrical-optical model and marked template matching, ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. III-7, pp. 75 - 82, [Online]. Available: https://www.isprs -ann-photogramm-remotesens-spatial-inf-sci.net/III-7/75/2016/
O. Csillik, J. Cherbini, R. Johnson, A. Lyons, and M. Kelly, Identification of citrus trees from unmanned aerial vehicle imagery using convolutional neural networks, Drones, vol. 2, no. 4, [Online]. Available: http://www.mdpi. com/2504-446X/2/4/39
K. Shivanna and R. Tandon, Seedling recruitment, in Reproductive Ecology of Flowering Plants: A Manual. Springer, 2014, pp. 145 - 162. DOI: 10.1007/978-81-322-2003-9_12
J. A. López, E. I. Verdiguier, L. G. Chova, J. M. Marí, J. R. Barreiro, G. C. Valls, and J. C. Maravilla, Land cover classification of vhr airborne images for citrus grove identification, ISPRS journal of photogrammetry and remote sensing, vol. 66, no. 1, pp. 115 - 123.
M. Zortea, M. Macedo, A. Britto Mattos, B. C Ruga, and B. H Gemignani, Automatic citrus tree detection from uav images based on convolutional neural networks, in 2018 31th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), 11 2018.
D. K. Nkemelu, D. Omeiza, and N. Lubalo, Deep convolutional neural network for plant seedlings classification, CoRR, vol. abs/ 08404, 2018. [Online]. Available: http://arxiv.org/abs/1811.08404
H. S. Abdullahi, R. Sheriff, and F. Mahieddine, Convolution neural network in precision agriculture for plant image recognition and classification, in 2017 Seventh International Conference on Innovative Computing Technology (Intech), Ieee, Londrés, 2017, pp. 1 - 3.
T. Ojala, M. Pietikäinen, and D. Harwood, A comparative study of texture measures with classification based on featured distributions, Pattern recognition, vol. 29, no. 1, pp. 51 - 59, DOI: 10.1016/0031-3203(95)00067-4
R. M. Haralick, K. Shanmugam, and I. Dinstein, Textural features for image classification, IEEE Transactions on Systems, Man, and Cybernetics, vol. SMC- 3, no. 6, pp. 610 - 621, Nov DOI: 10.1109/tsmc.1973.4309314
S. Sural, G. Qian, and S. Pramanik, Segmentation and histogram generation using the hsv color space for image retrieval, in Proceedings. International Conference on Image Processing, vol. IEEE, 2002. DOI: 10.1109/icip.2002.1040019
Mohtadi Ben Fraj, In depth: Parameter tuning for KNN, https://medium.com/p/in-depth-parameter-tuning-for-knn-4c0de485baf6, Accessed: 2019 - 05.
In-depth: Parameter tuning for random forest, https://link.medium.com/QxBb1qVBRW, Accessed: 2019 - 05.
In-depth: Parameter tuning for SVC, https://link.medium.com/acuwGvW83W, Accessed: 2019 - 05.
H. Dongyuan and C. Xiaoyun, Tuning svm hyperparameters in the primal, in 2010 Second International Conference on Computational Intelligence and Natural Computing, vol. IEEE, 2010, pp. 201 - 204. DOI: 10.1109/cinc.2010.5643857
S. Bernard, L. Heutte, and S. Adam, Influence of hyperparameters on random forest accuracy, in International Workshop on Multiple Classifier Systems. Springer, 2009, pp. 171 - DOI: 10.1007/978-3-642-02326-2_18
N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, Smote: synthetic minority over-sampling technique, Journal of artificial intelligence research, vol. 16, pp. 321 - 357, DOI: 10.1613/jair.953
H. He, Y. Bai, E. A. Garcia, and S. Li, Adasyn: Adaptive synthetic sampling approach for imbalanced learning, in 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence). IEEE, 2008, pp. 1322 - 1328. DOI: 10.1109/ijcnn.2008.4633969
A. Fernández, S. Garcia, F. Herrera, and N. V. Chawla, Smote for learning from imbalanced data: progress and challenges, marking the 15-year anniversary, Journal of artificial intelligence research, vol. 61, pp. 863 - 905, 2018.
Publicado
09/09/2019
Como Citar
KOBAYASHI, Felipe Kawashita; MATTOS, Andrea Britto; MACEDO, Maysa M. G.; GEMIGNANI, Bruno H.
Citrus Tree Classification from UAV Images: Analysis and Experimental Results. In: WORKSHOP DE VISÃO COMPUTACIONAL (WVC), 15. , 2019, São Bernardo do Campo.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2019
.
p. 31-36.
DOI: https://doi.org/10.5753/wvc.2019.7624.
