Models for crop yield estimation using multi-temporal UAV-based remote sensing imagery

点击翻译按钮获取中文摘要

Precision farming is a new agriculture management approach based on observation and measurement to respond to inter and intra-field variability in crops. Empirical crop yield estimation models have been developed using satellite remote sensing technique. However, currently, there are only very few studies on the use of Unmanned Aerial Vehicle (UAV) systems for crop yield estimation. As an innovative technology and an inexpensive and more reliable alternative to airborne and satellite remote sensing, A&L has developed hardware and software for UAV system in order to provide crop monitoring and yield prediction service for its clients. This project aims to collaborate with A&L Canada Laboratories Inc. to develop models for crop yields estimation using multi-temporal UAV-based remote sensing imagery. The objective of this project is to develop methodology/procedures for crop yield prediction using the UAV images acquired by the A&L newly developed multispectral sensor. The proposed methodology includes (1) UAV image collection, image calibration and processing; (2) mapping of canopy Nitrogen based on UAV images and plant samples, data collection on soil properties and weather conditions; (3) Development of crop yield prediction models; (4) Comparison of yield prediction results using different combinations of dates, phenological stages and different input variables. The final results and algorithms of this project will be immediately integrated into a service A&L intends to offer to the Agriculture industry in the near future. A&L will be able to generate revenue from hardware sales as well as ongoing service sales for interpretations which will be unique to their sensor technology. The research and findings will directly benefit the Canadian agricultural sector. Efficiencies found in production agriculture directly benefit not only the producers but the entire food supply chain for the domestic and export market.