Digital Imaging System for High-Throughput Plant Phenotyping Using Raspberry Pi Computers

Digital imaging sensors can perform various functions via utilizing high-throughput phenotyping (HTP). The HTP platform using advanced sensors facilitates more robust and efficient data collection than conventional phenotyping. It allows accurate, rapid, and nondestructive measurement of plant morphometrics and physiological parameters representing plant health. By combining smart sensing systems, it is possible to characterize crop growth parameters at high spatial, spectral, and temporal resolution. During this study, we found the development of an Internet-of-Things (IoT)-based sensor system using low-cost Raspberry Pi (RPi) computers (≤$300) could help plant breeders and field agronomists to monitor plants characteristics over time. Its utility was demonstrated using kochia (Bassia scoparia) plants as a model weed species. The prominent phenotypic traits studied using RPi based indoor imaging setups include vegetation cover, canopy structure, plant density, leaf area and foliar greenness. The setup used to capture images was significantly cheaper than the other alternatives. With the IoTs advancement, it is needed to transfer the developed tools and methodologies for adoption by breeding programs and precision agriculture industry.