For the rapid detection of sesame oil flavoring, a real-time and accurate detection of sesame oil flavoring system is developed. The system mainly consists of the data acquisition part and data processing components. The data acquisition parts include gas sensor and the host computer. Data processing part include a three-layer BP neural network which is trained in MATLAB. During the experiment, the preparations of sesame oil flavoring samples have eight different proportions, and the standard samples are for the neural network training and test samples are verified. The system provides a basis method for the rapid detection of sesame oil flavoring.