Morphological image analysis for the detection of water stress in potted forsythia

The purpose of this work was to study how artificial vision could indicate a modification of the state of a plant on the basis of shape analysis methods. In order to do this, morphological indicators were defined, allowing a diagnosis in the case of plants where water deficits were fairly high. This study was conducted on images of forsythias analysed on a horizontal plane. The main section of this paper concerns the identification of features indicating the entire morphological development of plants under water stress conditions. Several methods were developed and tested. Three methods made it possible to accurately define a threshold above which the plant can be considered stressed. A physiological study of the water status of the plant was undertaken at the same time in order to validate the results obtained by artificial vision. (C) 2004 Silsoe Research Institute. All rights reserved. Published by Elsevier Ltd.