Abstract: The nondestructive inspection of sugar content (SC) and firmness of navel oranges was discussed using laser Raman spectroscopy. After analyzing Raman spectra of navel oranges obtained by Laser Raman spectroscopy, the authors can obtain four eigenvalues of SC and firmness of navel oranges. The three-layer BP neural network was set up and used to predict the firmness and SC of navel oranges, and the eigenvalues were the parameters of input of BP neural network model. The results showed that error variances of SC and firmness between predicted values and experimental measurements were 0.0656 and 0.0062. It is feasible to detect fruit quality nondestructively using laser Raman spectrum technology.