Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/20/2011
Publication Date: 6/20/2011
Citation: Wiles, L. 2011. Software to quantify and map vegetative cover in fallow fields for weed management decisions. Computers and Electronics in Agriculture. 78:106-115. Interpretive Summary: Researchers measure vegetative ground cover, the percentage of the ground covered by live vegetation, using image analysis of color digital images. A map of vegetative cover can be created if GPS coordinates are collected for image locations. Crop growers could use a map of vegetative cover of a crop to target the use of inputs by identifying areas in a field where the crop is stressed or weeds are a problem. A system of readily available consumer products and free software was developed for growers to map weed cover in fallow fields. Images are collected from a moving vehicle without a connection between the camera and GPS receiver. Users with minimal GIS expertise can create a map of vegetative cover and then are able to select a location on the map to view the color digital image and determine important characteristics like plant species and size. We implemented the system for $2500 and have been able to create a map from 1200 images and 5000 GPS coordinates in 30 minutes. The system works with a variety of consumer cameras and GPS receivers and could be modified to create maps to help public land managers detect areas that are susceptible erosion or the spread of invasive weeds and help livestock managers monitor rangeland health and prevent overgrazing.
Technical Abstract: Mapping weed cover during the fallow period of dryland crop rotations would be valuable for weed management in subsequent crops and could be done with low cost color digital cameras, however most managers lack the specialized software and expertise needed to create a map from the images. We developed a system of software to quantify weed cover in fallow fields in digital images and to simplify and automate the most challenging tasks that non-GIS professionals confront in creating and using maps derived from a large number of images. A GIS file of image locations is created with inexpensive consumer software. Images are classified, a GIS file is generated and the map is displayed in a simple GIS viewer with free software we developed. A map can be generated from 1000 images and 5000 GPS coordinates in 30 minutes, including image classification. The classified and original images for all locations can be viewed together easily from the map application. The accuracy of estimating weed cover was evaluated using images collected in 15 fields under natural light with a consumer grade camera mounted on an ATV driving 8 to 11 km hr-1. Weed cover was estimated with 96% accuracy for images regardless of the amount of crop residue, unless part of the image was shaded by the camera. In those images, accuracy was 90% or better. This system will work with many professional and consumer digital cameras and GPS units and the classification algorithm can be easily modified for other applications.