Abstract
The image processing system is composed of a thermal camera with a 8-14 μm band pass filter, video memory with A/D, D/A, IC memory (64 K Byte), and a timingcontroller and computer with 32 kW cores and a 2.5 MW disc memory. Data on the input images is processed by a FORTRAN, JIS 7000. After the processing, leaf distribution and the binary pattern of leaf temperature, respectively, are passed to colour display and to the printer. The temperature determination has an accuracy of about 0.1°C through out the total system after adjustments for leaf reflectence, emissibility and resolution.
Distribution of the leaf temperature in both tobacco and sunflower were measured by the image processing system under various plant environments. Binary images gained by the threshold leaf temperature play an important role in image analysis for plant physiology. We concluded that this image processing procedure is one of the most effective methods for investigating leaf information in plant growth.
