Abstract: Salted duck eggs are a type of traditional Chinese pickled delicacy, and moisture and lipid content are important indexes for evaluating the quality during processing. This study used a hyperspectral imaging (HSI) system in conjunction with chemometrics to investigate the content change and distribution of moisture and lipid during different salting stages of duck eggs. The HSI was used to obtain reflectance spectral information of salted duck eggs in the 432~961 nm wavelength range. To minimize the noise in spectral signals, three preprocessing methods including Savitzky-Golay smoothing (SG), Gauss filter smoothing (Gauss), and standard normal variation (SNV) were used. The competitive adaptive reweighted sampling (CARS) was used to select the optimal wavelengths for predicting moisture and lipid content, and then the partial least squares regression (PLSR) and artificial neural network (ANN) methods were used to predict moisture and lipid content quantitatively. Results showed that ANN model could exhibited a better performance in predicting moisture and lipid content with coefficients of determination of the protein moisture, yolk moisture and yolk lipid of 0.9306, 0.9552 and 0.8896 respectively. Finally, the ANN model was used to create a distribution map of moisture and lipid content in the profile of salted duck eggs. The visualization distribution maps successfully display the distribution of moisture and lipid content during different salting periods. This study could lead to the development of a rapid detection method based on spectrum technology for production and processing of salted duck egg industry.