Remote Sensing Analytics

Since the beginning, AggieAir researchers have advanced the use of remote sensing technology coupled with machine learning approaches to develop highly accurate classification methods. This is illustrated in the use of AggieAir imagery and machine learning approaches to classify invasive vegetation in wetlands, for example. Some of AggieAir's earliest work tracked the spread of Phragmites australis, in invasive reed causing serious damage to North American wetlands, using remote sensing techniques and machine learning. This research also resulted in a new understanding about the rate at which clones of Phragmites spread in dense, genetically identical patches.

In addition AggieAir remote sensing technology has seen the successful classification of submerged aquatic vegetation in a high-mountain lake, providing valuable information for management efforts to monitor and control the spreading of invasive milfoil.

Scientific-grade data generated through the application of AggieAir technology and protocols has led to a better understanding biophysical processes whose characteristics can be measured through the acquisition and analysis of high-resolution, multispectral imagery.

Below are a few examples:

• AggieAir researchers are involved in a multi-year partnership with the agricultural Research Service in USDA and E&J Gallo Winery to develop better models for remote sensing of water consumption in wine grape vineyards.
• AggieAir has developed machine learning models that uses AggieAir imagery to estimate surface and root-zone soil moisture in irrigation fields and to optimize the application of irrigation water.
• AggieAir imagery and machine learning approaches have been used to remotely sense crop tissue chlorophyll content.
• AggieAir have begun to research new analytic approaches for upscaling and downscaling methods to reconcile high-resolution UAV data with geographically coarse-scale satellite data.

The current state of UAV-based remote sensing technology, especially with respect to commercial applications in agricultural management leaves much to be desired, particularly in regards to scientific content. AggieAir researchers have begun to advocate the identification of scient-based standards for agricultural remote sensing.

References:

• Mapping ET at High Resolution Using AggieAir Airborne Multispectral Imagery
• Estimating chlorophyll with thermal and broadband multispectral high resolution imagery from an unmanned aerial system using relevance vector machines for precision agriculture
• Optimal Irrigation Water Allocation for a Center Pivot Using Remotely Sensed Data
• Topsoil moisture estimation for precision agriculture using unmanned aerial vehicle multispectral imagery
• Assessment of optimal irrigation water allocation for pressurized irrigation system using water balance approach, learning machines, and remotely sensed data
• Assessment of Surface Soil Moisture Using High-Resolution Multi-Spectral Imagery and Artificial Neural Networks
• Spatial Root Zone Soil Water Content Estimation in Agricultural Lands Using Bayesian-Based Artificial Neural Networks and High- Resolution Visual, NIR, and Thermal Imagery
• Spatial Scale Gap Filling Using an Unmanned Aerial System: A Statistical Downscaling Method for Applications in Precision Agriculture
• Life on the edge: reproductive mode and rate of invasive Phragmites australis patch expansion
• The remote sensing data from your UAV probably isn’t scientific, but it should be!
• Use of high-resolution multispectral imagery acquired with an autonomous unmanned aerial vehicle to quantify the spread of an invasive wetlands species
• Tracking Phragmites australis expansion in the Bear River Migratory Bird Refuge using AggieAir^TM aircraft data
• Fusion of remotely sensed data for landcover classification using multi-class relevance vector machine