This study explored the quantitative detection of hazelnut oil adulteration. A portable laser Raman spectrometer was used on 180 adulterated samples of hazelnut oil mixed with corn oil, walnut oil, and flaxseed oil. The acquired spectra were divided into calibration and validation sets in a 3∶1 ratio. Qualitative analysis was performed using principal component analysis and quantitative detection of adulterated samples was achieved by establishing a partial least-squares regression. The experiment yielded three binary pseudo samples of hazelnut oil mixed with flaxseed oil, walnut oil, and corn oil, respectively. The corresponding correlation coefficients were 0.9894, 0.9872, and 0.9688; the root mean square errors on calibration were 0.0037, 0.0098, and 0.0121; the root mean square errors on prediction were 0.0114, 0.0126, and 0.0190; and the relative analysis errors were 9.707, 8.848, and 5.662. The difference in the model parameters of the three kinds of adulterated samples was reasonably explained. The results show that the proposed system has excellent predictive performance for quantitative detection of hazelnut oil adulteration. The system can be used for simple, rapid, and nondestructive quantitative detection of hazelnut oil adulteration.