Odedeyi, Temitope; Oyinlola, Adeoluwa; (2025) Enhancing RF Reflectometry for Crop Quality Sensing Using AI-Based Frequency Selection. In: 2025 IEEE BioSensors Conference (BioSensors). IEEE: San Diego, CA, USA.

Preview

Text Enhancing_RF_Reflectometry_for_Cassava_Quality_Sensing_Using_AI_Based_Frequency_Selection_Camera_Ready (1).pdf - Accepted Version Download (1MB) | Preview

Abstract

This study explores the integration of machine learning to enhance the predictive accuracy of RF reflectometry for crop quality sensing, specifically focusing on dry matter and starch content estimation in cassava. To enhance sensing accuracy, we implement AI-driven frequency selection using iterative linear regression and random forest models. A dataset of RF measurements spanning 30 kHz to 200 MHz was analysed to identify the most predictive frequency combinations. Results demonstrate that a single-frequency regression achieves an R 2 of 0.64, while iterative multiple linear regression for 2-frequency combinations yields an R 2 of 0.71. However, the ML techniques results show an R 2 of 0.67 for single frequency modelling and 0.78 for two frequency combinations. This would enhance the application and utility of crop quality sensors based on RF reflectometry measurement.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as CSVJSONXML

Archive Staff Only

View Item