Phenotypic analysis of fruit crops by water stress - Focused onthermal image analysis

摘要

Phenomics is a field of biology that systematically interprets all phenotypes which are shown in cells, tissues, organs, life forms, etc. Recently, it has become increasingly meaningful as a technology to develop plant phenotype that combines ICT, bio, and optical technology. Now, research about phenotypic analysis is essential in the field of smart farm and among them, leaf temperature of crops provides important physiological information for crop management such as moisture content, water stress and disease. By measuring plant temperature changes due to the opening and closing of crop pores, the response of crops to various stresses can be analyzed. Conventional contact sensors which are used to measure temperature have many disadvantages. For example, contact sensors are attached directly to the leaves, causing crop scars and also impossible to measure the temperature over the entire range of the leaves. Non-contact measurement using infrared thermal imaging requires temperature correction depending on the ambient environment and the type of crop, but it is useful for detecting the temperature of the crop and can be measured more accurately and precisely. In this study, leaf temperature information of crops is extracted after correcting temperature information which are acquired in thermal images. Using the extracted temperature, we confirmed the changes of the leaf temperature with time and the relationship between the leaf temperature and the amount of evapotranspiration, the leaf temperature and the water tension. Peach crops are used. After different soil water tension treatments, the temperature, humidity, illumination, water tension, evapotranspiration, leaf temperature, thermal imaging data were obtained. We analyzed the thermal emissivity of the crop by extracting the leaf temperature information of the crops measured by the thermal images and corrected the temperature information of the thermal images using this. After obtaining the spatial information of the leaves using the temperature-corrected thermal images, the temperature information of the leaves of each tree was extracted through image processing. The relationship between moisture stress and leaf temperature was analyzed by checking the relationship between the leaf temperature and the amount of evapotranspiration, the leaf temperature and water tension.