Embedded Vision AI for UAV-Based Weed Detection in Sugarcane Field Using Semantic Segmentation
Resumo
Weed infestation in sugarcane increases herbicide use, costs, and environmental pressure. This paper benchmarks 394 custom dense-prediction Semantic Segmentation variants across nine architectures (U-Net, PSPNet, FPN, SegNet, DeepLabV3+, FCN-ASPP, LinkNet, HRNet, SegFormer) for onboard UAV-based weed detection using high-resolution RGB images from commercial fields annotated into four foreground classes plus background. A rapid workflow was developed combining field validation, aerial imaging, manual polygon annotation, and tiled partitioning (128–224 px). Beyond standard metrics, we introduced lightweight inference-time uncertainty diagnostics to filter miscalibrated models. The pipeline enable practical edge for precision agriculture.
Palavras-chave:
semantic segmentation, edge AI, UAVs, precision agriculture, uncertainty diagnostics, TensorRT
Referências
Badrinarayanan, V., Kendall, A., and Cipolla, R. (2017). Segnet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(12):2481–2495.
Chaurasia, A. and Culurciello, E. (2017). Linknet: Exploiting encoder representations for efficient semantic segmentation. In 2017 IEEE Visual Communications and Image Processing, pages 1–4. IEEE.
Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., and Adam, H. (2018). Encoder-decoder with atrous separable convolution for semantic image segmentation. In Proceedings of the European Conference on Computer Vision (ECCV), pages 801–818. Springer.
Gao, J., Liao, W., Nuyttens, D., Lootens, P., Alexandersson, E., and Pieters, J. (2024). Cross-domain transfer learning for weed segmentation and mapping in precision farming using ground and UAV images. Expert Systems with Applications, 237:122980.
Garibaldi-Márquez, F., Flores, G., and Valentin-Coronado, L. M. (2023). Segmentation and classification networks for corn/weed detection under excessive field variabilities. In Pattern Recognition, pages 135–146. Springer Nature Switzerland.
Guo, C., Pleiss, G., Sun, Y., and Weinberger, K. Q. (2017). On calibration of modern neural networks. Proceedings of the 34th International Conference on Machine Learning (ICML), 70:1321–1330.
He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778.
Hu, J., Shen, L., and Sun, G. (2018). Squeeze-and-excitation networks. Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7132–7141.
Huang, G., Liu, Z., Van Der Maaten, L., and Weinberger, K. Q. (2017). Densely connected convolutional networks. Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4700–4708.
Koubaa, A., Ammar, A., Abdelkader, M., Alhabashi, Y., and Ghouti, L. (2023). AERO: AI-enabled remote sensing observation with onboard edge computing in UAVs. Remote Sensing, 15(7):1873.
Lin, T.-Y., Dollár, P., Girshick, R., He, K., Hariharan, B., and Belongie, S. (2017). Feature pyramid networks for object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2117–2125.
Machidon, O. M., Krašovec, A., Machidon, A. L., Pejovic, V., et al. (2023). AgriAdapt: Towards resource-efficient UAV weed detection using adaptable deep learning. In Proceedings of the 1st International Workshop on Advances in Environmental Sensing Systems for Smart Cities.
Maldaner, L. F., Molin, J. P., Canata, T. F., and Martello, M. (2021a). Monitoring weed infestation in sugarcane fields using uav-based multispectral imagery. Biosystems Engineering, 212:45–57.
Maldaner, L. F., Molin, J. P., Canata, T. F., and Martello, M. (2021b). Sugarcane yield mapping using high-resolution imagery and machine learning. Precision Agriculture, 22:1671–1691.
Martins, J. A. C., Higuti, A. Y. H., Pellegrin, A. O., et al. (2024). Assessment of UAV-based deep learning for corn crop analysis in midwest brazil. Agriculture, 14(11):2029.
Milioto, A., Lottes, P., and Stachniss, C. (2018). Real-time semantic segmentation of crop and weed for precision agriculture robots leveraging background knowledge in cnns. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages 2229–2235. IEEE.
Mohsan, S. A. H., Khan, M. A., Noor, F., Ullah, I., and Alsharif, M. H. (2023). Towards the unmanned aerial vehicles (uavs): A comprehensive review. Drones, 7(6):391.
Mohsan, S. A. H., Othman, N. Q. H., Li, Y., Alsharif, M. H., and Khan, M. A. (2022). Unmanned aerial vehicles (uavs): Practical aspects, applications, open challenges, security issues, and future trends. Intelligent Service Robotics, 16:109–137.
Moriya, É. A. S., Imai, N. N., Tommaselli, A. M. G., Berveglieri, A., and Honkavaara, E. (2023). Detection of weeds in sugarcane using uav-based hyperspectral imaging and deep learning. Remote Sensing, 15(9):2322.
Mukherjee, A., Misra, S., and Raghuwanshi, N. S. (2019). A survey of unmanned aerial sensing solutions in precision agriculture. Journal of Network and Computer Applications, 148:102461.
Mukherjee, A., Misra, S., Raghuwanshi, N. S., and Mitra, S. (2023). A comprehensive survey of uav-based precision agriculture: Current status and future perspectives. IEEE Access, 11:52626–52662.
Portz, G., Molin, J. P., and Jasper, J. (2023). Precision agriculture in sugarcane: A review on sensing technologies for crop and soil monitoring. Precision Agriculture, 24:1150–1181.
Shelhamer, E., Long, J., and Darrell, T. (2017). Fully convolutional networks for semantic segmentation. In IEEE Transactions on Pattern Analysis and Machine Intelligence, volume 39, pages 640–651.
Tan, M. and Le, Q. V. (2019). Efficientnet: Rethinking model scaling for convolutional neural networks. In Proceedings of the International Conference on Machine Learning (ICML), pages 6105–6114.
Xie, E., Wang, W., Yu, Z., Anandkumar, A., Alvarez, J. M., and Luo, P. (2021). Segformer: Simple and efficient design for semantic segmentation with transformers. In Advances in Neural Information Processing Systems (NeurIPS), volume 34, pages 12077–12090.
Xie, S., Girshick, R., Dollár, P., Tu, Z., and He, K. (2017). Aggregated residual transformations for deep neural networks. Proceedings of the IEEE conference on computer vision and pattern recognition, pages 1492–1500.
Zhao, H., Shi, J., Qi, X., Wang, X., and Jia, J. (2017). Pyramid scene parsing network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2881–2890.
Chaurasia, A. and Culurciello, E. (2017). Linknet: Exploiting encoder representations for efficient semantic segmentation. In 2017 IEEE Visual Communications and Image Processing, pages 1–4. IEEE.
Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., and Adam, H. (2018). Encoder-decoder with atrous separable convolution for semantic image segmentation. In Proceedings of the European Conference on Computer Vision (ECCV), pages 801–818. Springer.
Gao, J., Liao, W., Nuyttens, D., Lootens, P., Alexandersson, E., and Pieters, J. (2024). Cross-domain transfer learning for weed segmentation and mapping in precision farming using ground and UAV images. Expert Systems with Applications, 237:122980.
Garibaldi-Márquez, F., Flores, G., and Valentin-Coronado, L. M. (2023). Segmentation and classification networks for corn/weed detection under excessive field variabilities. In Pattern Recognition, pages 135–146. Springer Nature Switzerland.
Guo, C., Pleiss, G., Sun, Y., and Weinberger, K. Q. (2017). On calibration of modern neural networks. Proceedings of the 34th International Conference on Machine Learning (ICML), 70:1321–1330.
He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778.
Hu, J., Shen, L., and Sun, G. (2018). Squeeze-and-excitation networks. Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7132–7141.
Huang, G., Liu, Z., Van Der Maaten, L., and Weinberger, K. Q. (2017). Densely connected convolutional networks. Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4700–4708.
Koubaa, A., Ammar, A., Abdelkader, M., Alhabashi, Y., and Ghouti, L. (2023). AERO: AI-enabled remote sensing observation with onboard edge computing in UAVs. Remote Sensing, 15(7):1873.
Lin, T.-Y., Dollár, P., Girshick, R., He, K., Hariharan, B., and Belongie, S. (2017). Feature pyramid networks for object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2117–2125.
Machidon, O. M., Krašovec, A., Machidon, A. L., Pejovic, V., et al. (2023). AgriAdapt: Towards resource-efficient UAV weed detection using adaptable deep learning. In Proceedings of the 1st International Workshop on Advances in Environmental Sensing Systems for Smart Cities.
Maldaner, L. F., Molin, J. P., Canata, T. F., and Martello, M. (2021a). Monitoring weed infestation in sugarcane fields using uav-based multispectral imagery. Biosystems Engineering, 212:45–57.
Maldaner, L. F., Molin, J. P., Canata, T. F., and Martello, M. (2021b). Sugarcane yield mapping using high-resolution imagery and machine learning. Precision Agriculture, 22:1671–1691.
Martins, J. A. C., Higuti, A. Y. H., Pellegrin, A. O., et al. (2024). Assessment of UAV-based deep learning for corn crop analysis in midwest brazil. Agriculture, 14(11):2029.
Milioto, A., Lottes, P., and Stachniss, C. (2018). Real-time semantic segmentation of crop and weed for precision agriculture robots leveraging background knowledge in cnns. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages 2229–2235. IEEE.
Mohsan, S. A. H., Khan, M. A., Noor, F., Ullah, I., and Alsharif, M. H. (2023). Towards the unmanned aerial vehicles (uavs): A comprehensive review. Drones, 7(6):391.
Mohsan, S. A. H., Othman, N. Q. H., Li, Y., Alsharif, M. H., and Khan, M. A. (2022). Unmanned aerial vehicles (uavs): Practical aspects, applications, open challenges, security issues, and future trends. Intelligent Service Robotics, 16:109–137.
Moriya, É. A. S., Imai, N. N., Tommaselli, A. M. G., Berveglieri, A., and Honkavaara, E. (2023). Detection of weeds in sugarcane using uav-based hyperspectral imaging and deep learning. Remote Sensing, 15(9):2322.
Mukherjee, A., Misra, S., and Raghuwanshi, N. S. (2019). A survey of unmanned aerial sensing solutions in precision agriculture. Journal of Network and Computer Applications, 148:102461.
Mukherjee, A., Misra, S., Raghuwanshi, N. S., and Mitra, S. (2023). A comprehensive survey of uav-based precision agriculture: Current status and future perspectives. IEEE Access, 11:52626–52662.
Portz, G., Molin, J. P., and Jasper, J. (2023). Precision agriculture in sugarcane: A review on sensing technologies for crop and soil monitoring. Precision Agriculture, 24:1150–1181.
Shelhamer, E., Long, J., and Darrell, T. (2017). Fully convolutional networks for semantic segmentation. In IEEE Transactions on Pattern Analysis and Machine Intelligence, volume 39, pages 640–651.
Tan, M. and Le, Q. V. (2019). Efficientnet: Rethinking model scaling for convolutional neural networks. In Proceedings of the International Conference on Machine Learning (ICML), pages 6105–6114.
Xie, E., Wang, W., Yu, Z., Anandkumar, A., Alvarez, J. M., and Luo, P. (2021). Segformer: Simple and efficient design for semantic segmentation with transformers. In Advances in Neural Information Processing Systems (NeurIPS), volume 34, pages 12077–12090.
Xie, S., Girshick, R., Dollár, P., Tu, Z., and He, K. (2017). Aggregated residual transformations for deep neural networks. Proceedings of the IEEE conference on computer vision and pattern recognition, pages 1492–1500.
Zhao, H., Shi, J., Qi, X., Wang, X., and Jia, J. (2017). Pyramid scene parsing network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2881–2890.
Publicado
19/10/2026
Como Citar
PEREIRA, Robson Rogério Dutra; GRAZIOSI, Sandro Elias; INOUE, Roberto Santos.
Embedded Vision AI for UAV-Based Weed Detection in Sugarcane Field Using Semantic Segmentation. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 23. , 2026, Cuiabá/MT.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 1292-1303.
ISSN 2763-9061.
DOI: https://doi.org/10.5753/eniac.2026.31055.
