Early Detection of Water Stress in Tomato Plants Based on Projected Plant Area

Abstract

Early detection of water stress in tomato plants is required for the precise irrigation control to produce high-brix tomato fruits. For this purpose, we have developed a new water stress detection technique based on the projected plant area calculated from digital color images captured by a commercially available, inexpensive, digital still camera. As a first step, the effectiveness of using the projected plant area of tomato plants as a water stress index was confirmed by measuring the projected plant area along with other plant physiological information, e.g. leaf temperature, water potential, transpiration rate, and photosynthetic rate. Next, an image processing algorithm for the automated calculation of the projected plant area from digital color images was established using the discriminant analysis method. Furthermore, the most effective measurement angle of the projected plant area for the early detection of water stress was examined and measurement at an angle of 90° was proved to be the most sensitive to water stress in the tomato plant, i.e. wilting. The features of this technique are the stability of the projected plant area as an index of water stress, the low cost of the measurement equipment, and easy installation of the system. These features suggest that this technique can be introduced to commercial greenhouses to detect water stress in tomato plants and to be used for irrigation control for the production of high-brix tomatoes.