Skip to main content
Fig. 5 | Plant Methods

Fig. 5

From: A comparison of ImageJ and machine learning based image analysis methods to measure cassava bacterial blight disease severity

Fig. 5

Comparison of the ImageJ and machine learning analyses of CBB infected leaves. A Representative images from each timepoint (4-, 6-, and 9- DPI) of a Xam WT (top row) and XamΔTAL20 (bottom row) water-soaked spots were selected, visually inspected, and compared. The original images show the water-soaked spots from the color corrected images without segmentation from the background. The “ImageJ” images show water-soaked spots manually segmented from background and overlaid onto the RGB image. The machine learning images shows water-soaked spots segmented from background and pseudo-colored. Scale bar = 0.5 cm. B Water-soaked area data generated by ImageJ or machine learning were paired by inoculation location and plotted for 4 DPI (left plot), 6 DPI (middle plot), and 9 DPI (right plot). Calculated p-values (F-Variance test) shown in the upper corner of plot. Red = ImageJ Blue = machine learning

Back to article page

Plant Methods

ISSN: 1746-4811