T. Kobayashi et al., Detection of rice panicle blast with multispectral radiometer and the potential of using airborne multispectral scanners, PHYTOPATHOL, 91(3), 2001, pp. 316-323
Rice reflectance was measured to determine the spectral regions most sensit
ive to panicle blast infection. Reflectance increased in the 430- to 530-,
580- to 680-, and 1,480- to 2,000-nm regions at the dough stage both in the
laboratory and the field as the percentage of diseased spikelets increased
. The wavebands of the greatest sensitivity were in the visible region, loc
ated near 485 and 675 nm. After the yellow-ripe growth stage, near-infrared
rather than visible reflectance responded to panicle blast infections. Rat
ios of rice reflectance were evaluated as indicators of panicle blast. R470
/R570 (reflectance at 470 nm divided by reflectance at 570 nm), R520/R675,
and R570/R675 decreased significantly as the incidence of panicle blast inc
reased at the dough stage. At the yellow-ripe stage, R550/R970 and R725/R90
0 were used to estimate panicle blast severity as measured in terms of the
percentage of diseased spikelets. According to the simulation that uses gro
und-based sensor data, airborne multispectral scanners may be effective in
detecting the occurrence of panicle blast using a band combination of 530-
to 570- and 650- to 700-nm regions at the dough stage.