The purpose of this study is to monitor the growth of rice on a weekly basis by multicopter. The data collected were used to 1) determine whether top-dressing was required, 2) assess the potential for lodging risk, 3) estimate yield, 4) create maps of rice growth for protein content estimation. The normalized difference vegetation index (NDVI) and green excess index (2G_RBi) were both suitable for use as monitoring indices, and their application revealed the following: 1) The standard deviation of 2G_RBi was thought to be useful for determining the timing of top-dressing. The timing of top-dressing application was estimated most effective 10-15 days after maximum standard deviations were recorded. Areas with poor growth could also be identified using NDVI of the non-productive tillering stages and areas where top-dressing needed to be applied could be identified. 2) To diagnose lodging, plant height was estimated using the differences between the digital surface model (DSM) before the field was prepared for planting and on the monitoring day, and the risk of lodging 14 days before heading was shown for the entire area. 3) Yield was highly correlated with NDVI of the heading stage, and yield maps were created using a yield estimation equation. 4) With regard to eating quality, a strong correlation was observed between the protein content of brown rice and NDVI values from the heading stage to the first half of the maturing stage(15 days after heading stage), and accurate maps of eating quality were created.

The monitoring of rice growth using a multicopter is both safe and cost effective for individual farmers. By producing objective data and maps for assessments of top-dressing, lodging risk, yield, and protein contents, the findings presented here were shown to be useful for the detailed management of crop growth in fields.

Semi-arid North-central Namibia is susceptible to poor crop harvests due to flood and drought. To mitigate the risk on the food security caused by flood and drought, rice cultivation could be introduced into the seasonal ponds (oondombe), which is one of topographical features in this region. The successful introduction of rice cultivation will depend on a comprehensive understanding of the hydrology of water storage in such oondombe; however, few data exist to support this study. We therefore undertook spatiotemporal monitoring of oondombe water-storage volume by integrating satellite remote sensing with measurements using structure-from-motion multiview stereo (SfM-MVS) analysis with an unmanned aerial vehicle (UAV). SfM-MVS is a recently developed technique that enables precise, simple, and inexpensive measurement of topography. First, we conducted UAV surveys at 16 oondombe to generate a regression relationship between oondombe water expanse and oondombe water-storage volume. Then we observed daily oondombe water expanse using several different sources of long-term satellite data (AMSR-E, AMSR2, MODIS, Landsat ETM＋) interpreted with the assistance of recent data-fusion technique (database unmixing). Finally, we applied a regression relationship to the satellite data measurements of water expanse and obtained estimates of oondombe water-storage volume for the period 2002-2015 at three test sites in Namibia. The estimated oondombe water-storage volume closely reflected seasonal change and year-to-year variation in flood and drought status. The accuracy of UAV measurements was up to error margins of several centimeters in the vertical direction and 10cm in the horizontal direction. Comparison among oondombe water-storage volume estimates with three different spatial resolutions revealed that measurement with an insufficient spatial resolution may lead to the overestimation of water-storage capacity. This study not only revealed valuable data about oondombe in this region but also proposed a new approach for spatiotemporal hydrological monitoring over wide areas that merits additional research.

This paper describes a low-cost and flexible vegetation classification system using a small and lightweight unmanned aerial vehicle (UAV). Compared with traditional remote sensing systems such as airplanes and satellites, small UAVs can obtain very high (a few cm) resolution aerial images. They are unaffected by cloud cover because they can fly at a very low altitude. In addition, owing to their small size and light weight, UAVs are highly portable, have a low operational cost, and can be operated safely. However vegetation classification based on traditional pixel-based classification approaches cannot work well when aerial images with cm-level resolution are used. Furthermore, the area of the aerial image captured by a UAV is limited to a small region due to its low-altitude flight. We propose a new method for determining vegetation type from aerial images obtained by a small UAV based on superpixel segmentation and machine learning techniques. We developed a method for creating a wide-area high-resolution mosaic image from multiple aerial images obtained by the UAV and onboard sensor data such as GPS and inertial sensors. Superpixel segmentation based on mean shift was then used to divide the mosaic image into small regions. Finally, the vegetation type was classified using the support vector machine (SVM) training and classifying process. We conducted our experiments at Yawata moor in Hiroshima Prefecture to classify moor vegetation. We evaluated the success rate of our proposed method based on the results of these experiments, and concluded that the small UAV system is both effective and useful for vegetation classification.

Rapid and all-weather detection of flood area is needed for monitoring and mitigating flood disasters. This paper addressed flood area detection by using ALOS-2 PALSAR-2 data acquired during the 2015 heavy rainfall disaster in Kanto and Tohoku area, Japan. We propose an approach to detect flood area by thresholding of amplitude image and interferometric coherence image for non-urban area and urban area, respectively. The PALSAR-2-derived flood areas are validated using the inundation map provided by the Geospatial Information Authority of Japan (GSI) and showed 75% accuracy and 0.51 kappa coefficient in flood/non-flood discrimination. Effectiveness of lower incidence angle (less than 40 degrees) and a high sensitive observation mode (6m resolution mode) for detecting non-urban flood are also demonstrated by a comparative study. Interferometric phase variation was revealed to be effective for detecting urban-area flood compared to conventional interferometric coherence. The results demonstrated the feasibility of PALSAR-2 for rapid flood monitoring and can be used as a reference for possible future flood disasters.