Accurate plot grid creation using high-res multispectral imagery and deep learning
This solution leverages RTK-equipped UAVs with multispectral sensors and advanced deep learning models to create accurate plot grids. It enhances automated open-field grid generation with high-resolution image processing and machine learning techniques.
Overview
This innovative solution utilizes RTK-enabled UAVs equipped with multispectral sensors in combination with advanced deep learning models to generate precise plot grids over open fields. The integration of high-resolution image processing techniques with deep learning algorithms enhances the accuracy and precision of automated grid generation tasks. This approach is particularly beneficial for applications in agriculture, construction, and environmental monitoring, where accurate mapping and monitoring of large areas are critical.
Technical specifications
- RTK-equipped UAVs: These drones achieve centimeter-level accuracy in mapping areas from 0.02 to 14 km², depending on flight height.
- Multispectral sensors: Capture detailed visual and spectral data to provide comprehensive analysis of field conditions.
- Deep learning models: Applied for regression, detection, and segmentation tasks, achieving over 65% in performance metrics.
- High-resolution image processing: Techniques such as cropping, up-sampling, and down-sampling ensure data integrity and align with DL model requirements.
- Grid creation algorithms: Tailored algorithms enhance the precision of plot grid generation from detected plots.
Technology readiness level
This technology is currently at TRL 5, indicating that it has been validated in a relevant environment. Further validation is planned to ensure maximum accuracy and mitigate potential errors through field trials and refinement of models and algorithms.
About Xi’an Jiaotong-Liverpool University
Xi’an Jiaotong‑Liverpool University is a Sino‑British joint university in Suzhou, operating as a research‑led, internationally oriented institution. Its main campus in the Suzhou Dushu Lake Science and Education Innovation District of Suzhou Industrial Park gives companies access to a dense innovation ecosystem and nearby advanced manufacturing. The Entrepreneur College in Taicang runs industry‑themed schools and an embedded co‑education model that aligns talent pipelines and applied R&D with partner firms in a nearby high‑tech zone. Research is supported by competitive national, provincial and municipal programs in China, including awards from the National Natural Science Foundation of China. A dedicated Research Engagement and Innovation Office enables collaboration, technology transfer and IP management.
delivered to your inbox.