🌱 Chinese research team utilizes ground-based NDVI cameras + machine learning to create a high-precision real-time monitoring system for corn phenology! 🌱
✅ Fusion of spectral and textural features to crack the traditional satellite data lag problem
✅ Stacked Integrated Learning (SEL) model for accurate identification of BBCH phenological stages
✅ Real-time monitoring of leaf age, plant height, leaf area index, cover, plant density
✅ All-weather high-frequency observation for real-time decision support for precision fertilizer/irrigation application
👉 Highlights Express:
✅ Dual-light camera: NDVI+RGB color weather camera data hits the field details directly, at least 30 days earlier than satellite
✅ Bi-dimensional fusion: texture + spectral bi-dimensional modeling, breaking through the single feature limitations
✅ Data open source: more than 5000 real-time multi-spectral images of crops open source use (please leave a message if you need), modeling methods can be adapted to multi-crops, currently being adapted to real-time monitoring of the climate of the southern tea branch citrus, the future is promising!
✅ Domestic instrument: Multi-spectral weathering camera produced by StarView (www.xingshitu.com) Supported the publication of high-level academic papers (CAS Region I, Impact Factor 6.3)
Figure 1 Thesis cover
Figure 2 Example of RGB color image (June 14, 2024, imaging time 19:03)
Figure 3 Example of a near-infrared image (June 14, 2024, imaging time 19:03)
🌽 Self-developed intelligent dual-optical camera reveals crop growth details from multiple spectral dimensions
🌽 Application extension: we can apply the multispectral camera to many fields such as crop growth monitoring, early warning of pests and diseases, etc. We are looking forward to cooperation!
🌾🚜 Link to the paper: https://doi.org/10.1016/j.atech.2025.100850
Comments are closed