UAV-Based Hyperspectral Remote Sensing for Precision Agriculture: Challenges and Opportunities
Abstract
Modern agricultural production relies on monitoring crop status by observing and measuring variables such as soil condition, plant health, fertilizer and pesticide effect, irrigation and crop yield. Managing all of these factors is a considerable challenge for crop producers. As such, providing integrated technological solutions that enable improved diagnostics of field condition to maximize profits, while minimizing environmental impacts, would be of much interest. Such challenges can be addressed by implementing remote sensing systems such as hyperspectral imaging to produce precise biophysical indicator maps across the various cycles of crop development. Recent progress in unmanned aerial vehicles (UAVs) have advanced traditional satellite-based capabilities, providing a capacity for high-spatial, spectral and temporal response. However, while some hyperspectral sensors have been developed for use onboard UAVs, significant investment is required to develop a system and data processing workflow that retrieves accurately georeferenced mosaics. Here we explore the use of a pushbroom hyperspectral camera that is integrated on-board a multi-rotor UAV system to measure the surface reflectance in 272 distinct spectral bands across a wavelengths range spanning 400-1000 nm, and outline the requirement for sensor calibration, integration onto a stable UAV platform enabling accurate positional data, flight planning, and development of data post-processing workflows for georeferenced mosaics. The provision of high-quality and geo-corrected imagery facilitates the development of metrics of vegetation health that can be used to identify potential problems such as production inefficiencies, diseases and nutrient deficiencies and other data-streams to enable improved crop management. Immense opportunities remain to be exploited in the implementation of UAV-based hyperspectral sensing (and its combination with other imaging systems) to provide a transferable and scalable integrated framework for crop growth monitoring and yield prediction. Here we explore some of the challenges and issues in translating the available technological capacity into a useful and useable image collection and processing flow-path that enables these potential applications to be better realized.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2017
- Bibcode:
- 2017AGUFM.B51A1772A
- Keywords:
-
- 0402 Agricultural systems;
- BIOGEOSCIENCES;
- 0422 Bio-optics;
- BIOGEOSCIENCES;
- 0465 Microbiology: ecology;
- physiology and genomics;
- BIOGEOSCIENCES;
- 0480 Remote sensing;
- BIOGEOSCIENCES