267-3 A New Java Program to Rapidly Quantify Several Turfgrass Parameters from Digital Images.
Tuesday, October 24, 2017: 2:05 PM
Marriott Tampa Waterside, Florida Salon I-III
Digital image analysis (DIA) is an effective tool to objectively measure green turfgrass coverage and color. Turfgrass DIA procedures were originally developed using SigmaScan (SS) software for which software-specific macros were published to automate the analyses of image batches. The SigmaScan software has not been updated since the early 2000s and suffers compatibility issues with newer operating systems, and operates relatively slow or crashes when analyzing high-resolution images. A stand-alone program is needed that is compatible with current operating systems and capable of quickly batch-analyzing high-resolution turfgrass images. TurfAnalyzer is a java-based application that will run on Windows, Mac, and Linux operating systems and is available for download at turfanalyzer.com. It will batch-analyze a folder of turf images for green coverage and color based on user-defined hue, saturation, and brightness thresholds. Across species and turf types DIA results from TA have been highly correlated (r > 0.99) with SS. Furthermore, TA is capable of analyzing high-resolution images and requires approximately 1% of the processing time relative to SS. Turf Analyzer also has a frame analysis feature for evaluating smaller turf areas, such as greenhouse pots, lysimeters, and divots. Turf density and uniformity may also be measured with TA. Density is evaluated by characterizing shadows within the image. Uniformity is evaluated by measuring color characteristics within a resampled and blurred image. TurfAnalyzer is capable of providing an overall turf quality score (1 - 9) as a weighted average of coverage, color, density and uniformity evaluations. A companion application that is capable of parsing and evaluating individual plots from a single aerial image will also be discussed.