Rubber-Tree Leaf Diseases Mapping Using Close Range Remote Sensing Images

Abstract

Currently, close-range remote sensing method using drone-based platform which payload compact sensor has been used for monitoring and mapping in the agriculture sector at large area. Thus, this study is deployed drone with a compact sensor to identify the rubber tree leaf diseases based on two groups of a spectral wavelength which are visible (RGB: 0.4 μm – 0.7 μm) and near infrared (NIR: 0.7μm – 2.0 μm), respectively. Spectral obtained from drone-based platform will be validated using ground observation handheld spectroradiometer. Eight types of rubber tree clones leaf at three different conditions (healthy, unhealthy and severe) were randomly selected within the 9.4-hectare experimental rubber plot, Rubber Research Institute of Malaysia (RRIM), Kota Tinggi, Johor whereby consist RRIM 2000 series, RRIM 3000 series, and PB series, respectively. Based on the result, quantitative analysis shows that the f-value is smaller than Critical-one tail for healthy, unhealthy while for severe the f-value is larger than Critical-one tail. The f-value is 2.887 < 4.283 (healthy), 0.002 < 0.264 (unhealthy) and 1.008 > 0.0526, respectively. Thus, this can be concluded that spectral and estimate is equal at the 0.05 significant levels. While qualitative analysis shows that each rubber clone tree diseases can be distinguished at the near infrared band for healthy, unhealthy, and severe, respectively.