Study On High-Spectral Characteristics Of Cotton Infected With Verticillium Wilt And Diagnosing Models By Spectrum

【Objective】In order to elucidated spectrum characteristics of cotton canopy and leaf infected with verticillium wilt , estimated them SL ( severity level )to provide theoretic foundation for further monitoring cotton verticillium wilt at large scale using airborne and airspace remote sensing as flats and reference for monitoring cotton other disease【Method】Used datas that were collected both cotton varieties infected with verticillium wilt of canopy and single leaf spectrum and SL of disease which was measured at cotton disease garden and field in different growth phases in the year 2005～2006, meanwhile, SL from canopy and single leaf infected with verticillium wilt were investigated. Through qualitatively and quantitatively analyzed the correlation between SL of disease and reflectance spectrum,derivative spectra of cotton canopy and single leaf infected with verticillium wilt , we set up spectrum identify and inversion model of SL on cotton canopy and single leaf infected with verticillium wilt.【Result】The results indicated that Spectrum characteristic on cotton canopy and single leaf infected with verticillium wilt had better regularity with the increase of SL in different periods and varieties. Compare with normal cotton leaf,spectrum reflectance of leaf infected with verticillium wilt was increased significantly in visible light region (400～700nm) and ear-infrared region (700～1300nm) with the increase of the SL ,and specially signification in blue- violet to red regions(525～680nm). Spectrum reflectance of canopy infected with verticillium wilt increased significantly in visible light red region (620～700nm) with the increase of the SL of disease, but it inverted in near-infrared region (700～1300nm) and compare with single leaf infected with verticillium wilt ,specially signification in 780～1300nm.when SL of disease got 25%, both cotton canopy and leaf infected with verticillium wilt all changed significantly ,so it could been used as a watershed and diagnosed index to recognize cotton canopy and leaf infected with verticillium wilt in early time. There were evident difference between first derivative spectrum characteristics and SL of verticillium wilt in different treatments , not only cotton canopy but single leaf infected with verticillium wilt all changed significantly in red edge ranges(680～780nm) with different SL of disease. After analyzed derivative spectra, we found that red edge swing on cotton canopy and single leaf infected with verticillium wilt all decreased, and red edge position all equal moved to the blue, those all manifested the especial spectrum characteristics of disease. The thesis indicated that 434～724nm and 909～1600nm were selected out as sensitive band region to single leaf SL of verticillium wilt. 1001～1110nm and 1205～1320nm were selected out as sensitive band region to canopy SL of verticillium wilt. and 698nm were selected out as characteristic band of single leaf SL of verticillium wilt., 806nm were selected out as characteristic band of canopy SL of verticillium wilt. Accoding to above fact, some spectrum identify and inversion SL models on cotton canopy and single leaf of verticillium wilt all reached the best significantly level.【Conclusion】The results suggested that spectrum characteristics on cotton canopy and leaf infected with verticillium wilt were obvious. Those model in which spectrum identify and inversion SL of modes of verticillium wilt on cotton canopy and single leaf ,the first derivative spectrum at 723nm and 731nm: FD723,y = -412.84x+4.5149 R2=0.8564; FD731,y = -509.42x+5.3279 R2=0.7147; could invert accurately the cotton canopy and single leaf infected with verticillium wilt , and they may be used to forecasting the position of cotton infected with verticillium wilt , quantitatively.