Water activity (aw) of dehydrated pineapple is one of the most important quality factors that must be determined in the routine operation of a factory. A non-destructive technique for detecting aw of dehydrated pineapples in the factory is required. Near infrared hyperspectral imaging (NIR-HSI) that has previously been shown to be a possible non-destructive, rapid, accurate and robust method was used in this study. The model for aw was established using partial least square regression (PLSR). Spectra in the wavelength of 935–1720 nm of samples were measured by using NIR-HSI and preprocessing methods tested before model establishment. The accuracy of the prediction model for aw gave a correlation coefficient of prediction (Rp) of 0.72 and root mean square error of prediction (RMSEP) of 0.0054. Results showed that NIR-HSI could possibly be used for determining aw of dehydrated pineapple nondestructively and could be incorporated into the production process for online grading in dehydration factories.