The Application of Neural Networks for the Anomaly Detection in Plant Development
Abstract
Machine learning and neural networks have become a common tool in tasks related to image processing and analysis. One of these tasks is to detect anomalies in images or abnormal images. Computer vision is increasingly helping to digitalize production and reveal patterns in agriculture. Technology allows a human to speed up the analysis, but it is not possible to completely replace an experienced specialist. In this paper, machine learning methods are used to detect abnormal sequences of plant images. The results obtained can be used in agricultural automation tasks, where one of the main means of monitoring the condition of plants is photo or video. The developed software package will be able to automatically detect problems with the health of plants or surrounding infrastructure, which will help to take appropriate measures in time. In this article, two neural network approaches to detecting anomalies in the process of plant development using images are implemented. Experiments with real data were done as well as a comparative analysis of the efficiency of methods in a specific task.
Full Text:
PDF (Russian)References
S. Festag и C. Spreckelsen, «Semantic Anomaly Detection in Medical Time Series,» т. 278, 2021.
J. E. Zhang, D. Wu и B. Boulet, «Time Series Anomaly Detection for Smart Grids: A Survey,» в 2021 IEEE Electrical Power and Energy Conference (EPEC), 2021.
G. Colò, «Anomaly detection for cyber security: time series forecasting and deep learning,» Int. J. Sci. Res. Math. Stat. Sci, т. 7, p. 40–52, 2020.
M. Abdallah, W. J. Lee, N. Raghunathan, C. Mousoulis, J. W. Sutherland и S. Bagchi, «Anomaly detection through transfer learning in agriculture and manufacturing IoT systems,» arXiv preprint arXiv:2102.05814, 2021.
W. Li, V. Mahadevan и N. Vasconcelos, «Anomaly detection and localization in crowded scenes,» IEEE transactions on pattern analysis and machine intelligence, т. 36, p. 18–32, 2013.
M. Ravanbakhsh, M. Nabi, E. Sangineto, L. Marcenaro, C. Regazzoni и N. Sebe, «Abnormal event detection in videos using generative adversarial nets,» в 2017 IEEE international conference on image processing (ICIP), 2017.
O. Ronneberger, P. Fischer и T. Brox, «U-net: Convolutional networks for biomedical image segmentation,» в International Conference on Medical image computing and computer-assisted intervention, 2015.
A. Geiger, D. Liu, S. Alnegheimish, A. Cuesta-Infante и K. Veeramachaneni, «TadGAN: Time series anomaly detection using generative adversarial networks,» в 2020 IEEE International Conference on Big Data (Big Data), 2020.
L. Ruff, R. Vandermeulen, N. Goernitz, L. Deecke, S. A. Siddiqui, A. Binder, E. Müller и M. Kloft, «Deep One-Class Classification,» в Proceedings of the 35th International Conference on Machine Learning, 2018.
I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville и Y. Bengio, «Generative adversarial nets,» Advances in neural information processing systems, т. 27, 2014.
S. Takase и N. Okazaki, «Positional Encoding to Control Output Sequence Length,» CoRR, т. abs/1904.07418, 2019.
U. Sara, M. Akter и M. S. Uddin, «Image quality assessment through FSIM, SSIM, MSE and PSNR—a comparative study,» Journal of Computer and Communications, т. 7, p. 8–18, 2019.
F. T. Liu, K. Ting и Z.-H. Zhou, «Isolation Forest,» 2009.
Refbacks
- There are currently no refbacks.
Abava Кибербезопасность MoNeTec 2024
ISSN: 2307-8162