As soil moisture prediction model with the whole near-infrared spectral regions is complex and the single-band prediction model is susceptible to environmental impact, a dual-band method for measuring soil moisture content (MC) rapidly was proposed in this study. A total of 116 soil samples were collected and the NIR reflectance spectra of all soil samples were measured. The spectral data were transformed to new spectral data with logarithmic transformation and logarithmic of reciprocal transformation. The spectral bands of the new spectral data and the original spectral data that is sensitive and insensitive to soil moisture content were achieved by correlation coefficient method. The models for prediction of soil MC were developed based on single linear regression (SLR) with the sensitive spectral band and insensitive spectral band. The results show that the prediction precision of the models was high and the highest correlation coefficient of the models reached 0.9913 with the root mean square error of prediction RMSEP of 1.1983%. Thus, it is concluded that the methods used in this paper are available methods for rapid detection of soil MC and also provides theoretical basis for developing a low-cost portable near-infrared moisture meter.