Click on “Download PDF” for the PDF version or on the title for the HTML version.


If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Predicting leaf water potential of potato using spectral reflectance indices

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:   No Citation available.
Authors:   R. Zakaluk, R. Sri Ranjan
Keywords:   leaf water potential, nitrogen, IHS transformation, chromaticity transformation, principal components, vegetation indices, potato, remote sensing, neural network, digital camera

The purpose of this research was to investigate the utility of a 5 mega pixel, RGB (red green blue) digital camera mounted on a 2.5 m telescopic pole to determine the leaf water potential () of potato plants in the field using artificial neural network (ANN) modeling. A randomized 45 x 45 m systematic grid sampling design was employed. Sample plots were measured at random to obtain leaf water potential, nitrate content, volumetric water content, and digital imagery. The imagery from all plots was radiometrically calibrated and classified to isolate green foliage from soils, flowers, shadows, and senescent leaves. Along with the RGB imagery, 6 image transformations and 9 vegetation indices were evaluated as input neuron candidates to determine leaf water potential using ANN modeling. Input neuron candidates with significant co-linearity, were transformed using (PCA) principal components analysis. A linear relationship between soil nitrate and foliage greenness via the G image band was found to be significant. Using a training dataset, 8 input neurons were used to create an ANN model to determine leaf water potential measures. The PCA, image ratios, and image transformations provided additional information not found in the RGB imagery. The results show that for the validation dataset, the predicted and measured leaf water potential is from common populations.

(Download PDF)    (Export to EndNotes)