Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : BIOEDUSCIENCE

Early Detection of Pathogenic Fungi Curvularia sp. on Rice Seeds (Oryza sativa) Based on Modified Infrared Image Analysis Anggraini, Novita; Nirwanto, Herry; Mujoko, Tri
BIOEDUSCIENCE Vol 9 No 3 (2025): BIOEDUSCIENCE
Publisher : Universitas Muhammadiyah Prof. Dr. Hamka

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22236/jbes/17568

Abstract

Background: Curvularia sp. is a seed-borne pathogenic fungus that can reduce rice plant productivity. At the same time, conventional seed health testing methods still rely on visual observation and require a relatively long incubation time. This study aims to evaluate the potential of infrared image analysis based on a modified imaging system for early detection of Curvularia sp. infection in rice seeds. Methods: Seed health testing was performed using the blotter test method. At the same time, image acquisition was performed with a digital microscope equipped with an infrared light source, and images were analyzed using pseudocoloring and RGB-based color segmentation. Results: The results showed differences in infrared signal intensity patterns in Curvularia sp.-inoculated seeds, which could be identified on the fourth day after inoculation, earlier than visual observation, which showed symptoms on the fifth day. Detection accuracy was calculated using a confusion matrix based on visual observation as the reference method, with a sample size of 50 seeds per observation day, yielding an average detection accuracy of 91% over seven days of observation. Conclusions: The modified infrared image analysis method has the potential to serve as an early detection method for Curvularia sp. infection in rice seeds, although its performance depends on the limitations of the imaging system and the validation method used.