Taking ponceau 4R and amaranth as an example, concentration predictionand kind identification of synthetic food colors by fluorescence spectroscopy andradial basis function neural networks are introduced. By using SP-2558multifunctional spectral measuring system, the fluorescence spectra were measuredfor solution of ponceau 4R and amaranth excited respectively by the light withthe wavelength of 300 and 400 nm. For each sample solution of ponceau 4R, 15emission wavelength values were selected. The fluorescence intensitycorresponding to the selected wavelength was used as the network characteristicparameters, and a radial basis function neural network for concentrationprediction was trained and constructed. It was employed to predict ponceau 4Rsolution concentration of the three kinds of samples, and the relative errors ofprediction were 1.42%, 1.44% and 3.93% respectively. In addition, for solution ofponceau 4R and amaranth, the fluorescence intensity corresponding to thefluorescence wavelength was used as the network characteristic parameters, and aradial basis function neural network for kind identification was trained andconstructed. It was employed to identify the kind of food colors, and theaccuracy is 100%. These results show that the method is convenient, fast, andhighly accurate, and can be used for the detection of synthetic food color infood safety supervision and management.neural networks; Concentration prediction; Kind identification; Food safety(JH08-18)和研究生创新项目(CX08B-088Z)资助