Proceedings Article | 10 October 2018
KEYWORDS: Agriculture, Atomic force microscopy, Image classification, Remote sensing, Satellites, Image processing, Image segmentation, Satellite imaging, Photography, Near infrared
The objective of this study is to monitor precision paddy field area and crop classification on a small town area. Monitoring of this agricultural sector is useful for evaluating the cultivated environmental condition, especially in mixed crop cultivated areas. This study evaluated the small town in Chungbuk, South Korea, using drone images for the years 2016. The images were acquired 15 times using the RGB and NIR sensors, respectively. Using the collaboration of drone images and smart farm map, we improved the accuracy of crop classification and crop identification. This study also evaluated the relationship between vegetation health and growth state of the study area using remote sensing and GIS techniques. The study area, Daeso town was produced by several crops such as rice, red pepper, corn, sweet potato, sesame, welsh onion, bean, tobacco, ginseng. This study acquired images using fixed wing drone from August 5 to August 6, 2016. It was found that the range of DN values of drone images by crop was different for each crop. The NDVI of this area showed the highest value in the rice with active vegetation and the value of 0 or more in the crops such as sweet potato, soybean, corn and grassland. An object-based technique is used for crop classification. The results showed that scale 250, shape 0.1, color 0.9, compactness 0.5 and smoothness 0.5 were the optimum parameter values in image segmentation. As a result, the collaborated method showed that the kappa coefficient was 0.85 and the overall accuracy of classification was 88.0 %. The result of the present study validates our attempts for crop classification using high resolution drone image and collaborated smart farm map (SFM) established the possibility of using such remote sensing techniques widely to access the difficulty of remote sensing data acquisition in agricultural sector.