Visualizing plant conditions via remote sensing has attracted attention as a strategy for improving crop cultivation methods under climate change. We developed a quantitative evaluation method for crop growth dynamics, but it is challenging to apply the method for wide-area evaluation. In this study, we measured canopy temperature and the spectral reflectance above the canopy to evaluate the crop stress condition; this method can be applied for wide area evaluation. We attempt to develop a method for understanding crop stress conditions during the growing season using these measurements and mathematical models and solve the challenge of wide-area evaluation using drones.

Crop production is highly vulnerable and unstable in the target regions of this study, Southeast Asia and Africa, not only because of poor soil nutrition but also because of high variability in the water environment. We attempt to quantify the interactions between nutrients and water environments in crop growth using non-destructive measurements and equation models. Based on the outcome, we will propose appropriate cultivation management techniques, such as time and amount of fertilizer application under different water environments. Specifically, we plan to establish evaluation methods for locally applicable cultivation environments and management methods by measuring leaf area and spectral reflectance non-destructively in field trials in Japan under various nutrient and water environments and verify these methods via field trials in the target regions.

Although we developed a simple evaluation method for crop growth, it was difficult to understand the physiological traits during the reproductive growth stage. In this study, we established a simple technique for evaluating the canopy structure of crops during the reproductive growth stage by quantifying the distribution of leaf mass and leaf quality. We used non-destructive stratified measurements during the reproductive stage and mathematical models to evaluate community structure and also attempted to detect cultivar differences in canopy productivity distribution using canopy photosynthetic models. Furthermore, we aim to establish a method for gathering crop-growth information throughout the growing season by integrating the method with a simple evaluation method for the established vegetative stage. We believe that this will enable yield predictions that consider the yield formation process and provide a detailed understanding of stress information.

Simple crop growth evaluations and yield predictions are required to improve crop cultivation efficiency under climate change conditions. We developed a crop-yield prediction method using non-destructive measurements and mathematical models. Drones have become widely used in recent years, and their use has made prediction methods simpler and more widespread. In addition, artificial intelligence (AI) is considered a tool for improving prediction accuracy. In this study, we attempt to develop a model for simple, wide-area, and highly accurate prediction of crop yield using drone measurements and AI technology. With mathematical models and multispectral and thermal images acquired via drone measurements during the growing season, we intend to examine different machine learning and deep learning methods to propose an optimal learning model for improving prediction accuracy.

In recent years, attempts have been made to “diagnose” crops in the field using infrared thermography, but it is unclear whether these diagnostic data are consistent with intracellular changes. Although we can “observe” in the laboratory the changes at the cellular level in organelles in three dimensions using electron microscopy, it is not suitable for evaluating crop plants or canopies because of the cellular scale. In this study, we used a two-dimensional imaging chlorophyll fluorescence spectrometer to measure stress damage at the tissue level, with the aim of linking three-dimensional analysis at the electron microscope level with the technique used in wide-area observation of stress damage. We aim to establish a method that can be applied for wide-area observations used in fields and for stress damage monitoring. We will establish a diagnostic imaging technique that integrates “observation” and “diagnosis” and compare it with the simple and non-destructive observation of stress damage in fields. Furthermore, we believe this will be a pioneering case study for understanding environmental stress damage over a wide area via the visualization of cellular disorders at the canopy level.

Spent coffee grounds are often discharged from coffee beverage factories, restaurants, coffee shops, fast-food restaurants, and households and are disposed of as waste. Moreover, phosphorus and potassium, which are discharged for pH adjustment during the production of lactic acid bacteria, are disposed of as waste. We are currently investigating the effective use of these wastes in plant cultivation. Specifically, we identified plants that could be easily cultivated with each material and examined the amount and frequency of use. These activities will be conducted geared towards achieving part of the SDGs.