Taking the air-dried samples of five soil types from middle and eastern China as test materials, the correlations of their organic matter content with the spectral reflectance of near-infrared (1000-2500 nm), and with the ratio index (RI), difference index (DI), and normalized difference index (ND) of the first derivative values of the reflectance between two bands were studied. Based on this, the key spectral indices and the quantitative models for estimating soil organic matter (SOM) content were developed. After corrected with Multiplicative Scatter Correction (MSC) and Savitzky-Golay (SG) smoothing methods, the spectral reflectance of near-infrared had an obviously high correlation with SOM, compared with the original spectral reflectance, while the corrected spectral indices of the first derivative values of the reflectance between two bands took the intermediate position. The correlation of the spectral indices with SOM was in the order of was DI > RI > ND, regardless the composition of the original spectral reflectance or the first derivative spectra. The DI of the reflectance of near-infrared between 1883 and 2065 nm corrected with MSC and SG smoothing methods [DI(CR1883, CR2065)] had the best linear correlations with SOM. The test of the monitoring model based on DI(CR1883, CR2065) with the independent datasets of SOM showed that the R2 and RMSE validation values were 0.837 and 4.06, respectively. Comparing with the results from the Partial Least Square (PLS) method, the monitoring model based on DI (CR1883, CR2065) was somewhat inferior. However, the DI(CR1883, CR2065) only needed two reflectance bands, and the monitoring model was simpler, being able to provide more available information for developing portable instruments, and a good spectral index for estimating SOM content.