| Cite this article: | 陆棚,刘明堂,李斌,等.基于HMM+LSTM算法的农作物数字孪生体生长模型设计[J].灌溉排水学报,0,():-. |
| lu peng,liu mingtang,li bin,et al.基于HMM+LSTM算法的农作物数字孪生体生长模型设计[J].灌溉排水学报,0,():-. |
|
| |
|
|
| This article has been：Browse 30Times Download 0Times |
|
|
|
| DOI： |
|
| Design of Crop Digital Twin Growth Model Based on HMM+LSTM Algorithm |
|
lu peng1, liu mingtang1, li bin1, li shihao1, wang changchun1, jiang enhui2
|
|
1.North China University of Water Resources and Electric Power;2.Yellow River Water Resources Research Institute
|
| Abstract: |
| 【Objective】In order to improve the intelligence, refinement, and efficiency of crop planting, this article takes the reticulated honeydew melon as an example to illustrate the digital simulation process of empowering crop planting with digital twin technology. 【Method】 Real time acquisition and prediction of the state of reticulated honeydew melons are achieved through high-definition cameras. A multidimensional data acquisition device is set up to associate the digital twin of reticulated honeydew melons with real entities, and a corresponding digital twin honeydew melon simulation model is established. Combined with Hidden Markov Model (HMM) and Long Short Term Memory (LSTM) algorithms, an intelligent inference and evolution model of reticulated honeydew melons with virtual control is realized. 【Result】 Five different growth cycle states of honeydew were constructed: seed, seedling, flower, leaf, and fruit. The overall recognition accuracy of honeydew twins generated by the HMM+LSTM model was high, with an accuracy of 85.3% for root and seedling cycles, 78.6% for leaf cycles, and 82.8% for average cycles. 【Conclusion】 The integration of computer vision, deep learning, and agricultural intelligence technology provides precise, efficient, and non-destructive visualization methods for modern agriculture, helping farmers develop more accurate and efficient management methods, and providing technical support for the application of digital twin technology in agricultural production. |
| Key words: Digital twin; Mesh honeydew melon; Hidden Markov; LSTM algorithm; Smart Agriculture |
|
|
|
|
|
|
