BO-CNN-BiLSTM deep learning model integrating multisource remote sensing data for improving winter wheat yield estimation
2024

Deep Learning Model for Estimating Winter Wheat Yield

Sample size: 93 publication 10 minutes Evidence: high

Author Information

Author(s): Zhang Lei, Li Changchun, Wu Xifang, Xiang Hengmao, Jiao Yinghua, Chai Huabin

Primary Institution: Henan Polytechnic University

Hypothesis

Can a deep learning model integrating multisource remote sensing data improve the accuracy of winter wheat yield estimation?

Conclusion

The BO-CNN-BiLSTM model can reliably estimate winter wheat yields, providing valuable insights for agricultural policymaking.

Supporting Evidence

  • The BCBL model achieved an R² of 0.81 and RMSE of 616.99 kg/ha.
  • SIF data significantly improved yield estimation accuracy across different models.
  • The model reliably identified critical stages of winter wheat yield formation.
  • Yield estimates were consistent with official statistics.
  • The model performed well even under varying climatic conditions.

Takeaway

This study created a smart computer program that helps farmers know how much wheat they will grow, using special satellite pictures and weather data.

Methodology

The study developed a BO-CNN-BiLSTM model that combines convolutional neural networks and bidirectional LSTM to analyze remote sensing and climate data for yield estimation.

Potential Biases

Potential bias may arise from the reliance on specific datasets that may not represent all climatic conditions.

Limitations

The model's performance may vary across different regions due to differences in feature distributions.

Participant Demographics

The study focused on winter wheat yields in Henan Province, China.

Statistical Information

P-Value

0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.3389/fpls.2024.1500499

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication