Quantitative assessment of mosaic disease severity based on digital image processing
Asmar Hasan, Widodo Widodo, Kikin Hamzah Mutaqin, S H Hidayat, Muhammad Taufik
Abstract
Abstract Quantitative assessment of plant diseases can be done relatively quickly and practically, especially when applying digital image processing. This paper discusses digital image potential for assessing mosaic symptoms caused by the Tobacco mosaic virus (TMV). Chili plants ( Capsicum annuum L.) were grown in pots and subjected to two treatments, i.e., non-inoculated (V0) and TMV inoculated (V1). Plant image recording using a Canon 750D camera with a kit lens was done once a week, starting when the mosaic symptoms are first visible, i.e., in the second-week post-inoculation. Recorded images in RAW format were first converted to TIFF, then subjected to further analysis using image processing applications, namely GIMP 2.8 and Fiji-ImageJ. Differences in the RGB profile of leaves given V0 and V1 treatment was observed. Non-inoculated leaves (V0) have a dark green color pattern, while TMV inoculated leaves (V1) tend to have a mixed color pattern of dark green and bright green. In general, this indicates a decrease in chloroplast’s ability to absorb light in diseased leaves, reducing the photosynthesis level. This preliminary experiment shows digital image processing’s potential for estimating the severity of mosaic disease with a high degree of accuracy and precision.