2018 年 38 巻 5 号 p. 451-457
In order to establish a method of drone-based remote sensing (drone-RS) in agriculture and environmental measurement, it is important to find a balance of price and performance of the near-infrared camera for the drone. We aim to establish a low-cost drone-RS system that can be used even for small-scale farming. We initially used a Yubaflex (BIZWORKS) near-infrared camera, however, a RedEdge (MicaSense) camera became available in 2017, making it a matter of interest to compare the two, including the ability of RedEdge to inherit data acquired by Yubaflex. In this report, we present the results of our comparison of these two types of near-infrared camera from the viewpoint of price and performance, including points to be noted in image processing. Our conclusions are the following: (1) Yubaflex-based normalized difference vegetation index (NDVI) values were lower than RedEdge-based NDVI values, however, NDVI values from both systems were mutually convertible. (2) When monitoring vegetation based on NDVI time series or NDVI images generated from multiple images, processing (using software or a program) was required to convert digital numbers (DNs) to radiance values.