In quantitative analysis of spectral data, noises and background interference always degrades the accuracy of spectral
feature extraction. Continuum-removal analysis enables the isolation of absorption features of interest, thus increasing
the coefficients of determination and facilitating the identification of more sensible absorption features. The purpose of
this study was to test continuum-removal methodology with Visual-NIR spectral data of tomato leaf. Through analyzing
the correlation between continuum-removal spectrum and nitrogen content, 15 characteristics parameters reflected
changing tendency of nitrogen content were chosen, which is at 335, 405, 500, 520, 540, 550, 560, 580, 620, 640, 683,
704, 720, 736 and 770 nm. Finally, the variance inflation analysis and stepwise regression method was used to develop
the prediction model of the nitrogen content of tomato leaf. The result showed that the predicted model, which used the
values of continuum-removal spectrum at 335 and 720nm as input variables, had high predictive ability, with R2 of 0.755.
The root mean square errors of prediction using a leave-one-out cross validation method were 0.513. These results
suggest that the continuum-removal spectroscopy analysis has better potential to diagnose tomato growth in greenhouse.