The early detection of Fusarium sp. infection in rice seeds is crucial for improving agricultural productivity and food security. Traditional methods like the Blotter Test, while effective, are time-consuming and require specialized personnel. This study explores the potential of thermal imaging technology to detect Fusarium sp. infections on rice seeds quickly and non-destructively. Rice seeds were inoculated with Fusarium sp. and incubated for seven days, during which surface temperatures were measured daily using the Fluke iSee TC01A thermal camera. The results showed that infected seeds exhibited significantly higher surface temperatures compared to control seeds, particularly from days 3 to 6 of incubation. Scatterplot analyses revealed clear temperature differences between infected and uninfected seeds, supporting the hypothesis that thermal imaging can serve as an early indicator of Fusarium infection. The study also demonstrated the high sensitivity and specificity of thermal imaging, particularly on days 2 to 4 of the incubation period. Logistic regression analysis confirmed the significant relationship between seed temperature and infection status, with prediction accuracy up to 91%. This research suggests that thermal imaging technology could replace traditional methods, offering a faster, more efficient approach for seed health monitoring in the agricultural industry.
Copyrights © 2026