To realize the rapid identification and detection of oil pollutants, a fluorescence spectrum detection system based on laser-induced fluorescence (LIF) technology is built, and the fluorescence spectra of three different oils, i.e., 0 ^# diesel, 95 ^# gasoline, and common kerosene, are obtained. Characteristic parameters are extracted from the spectral information. The standard deviation, center distance, and kurtosis coefficients of the fluorescence peak are taken as sensitive characteristic parameters for cluster analysis. Finally, the curve fitting method is used to quantitatively measure the mass concentration of samples. Experimental results show that the combination of LIF technology with the characteristic parameter extraction method and curve fitting method can be used for the qualitative and quantitative detection of different oil pollutants, which provides a new idea for their rapid identification and detection.