Non-Destructive Method for Moisture Content Sensing Inside a Rice Storage

Abstract

Rice grains represent the majority of worldwide consumed daily food, especially for most countries in Asia, where rice crops symbolize the feature of the local culture. However, as rice grains are naturally hygroscopic, the total values (quality and quantity) are degrading due to their varying level of moisture content. Currently, a sampling moisture sensing based on a single-point measurement is employed to monitor the moisture content level. In this scenario, the conventional method needs to be revised because it is very localized and only represents part of the moisture distribution inside the bulk grains. Besides, implementing several high-end technologies is considerably expensive for small-scale industries in developing countries. Therefore, this study has developed an RTI system in a prototype scale for a constructive moisture sensing method. RTI is a unique approach that reconstructs an image across the monitored WSN area by exploiting the attenuation of RF signals caused by the presence of targeted subjects. Five rice moisture profiles at the percentage of 15%, 20% and 25% were reconstructed using image reconstruction algorithms, LBP, FBP, NOSER and TR. This study analyses the effectiveness of the proposed method in both simulation and experimental studies. The results positively support the possibility of engaging the RTI technique to localize the moisture distribution in rice storage.