The automated classification and recognition of stellar spectra is animportant research for the spectra processing system of modern telescope survey project. For the spectra without flux calibration, the authors present an automated stellar spectra classification system to achieve two goals: one is the spectral class and spectral subclass classification, and the other is luminosi tytype recognition. The system is composed of three units: (1)continuum normalization method based on wavelet technique; (2)non-parameter regression method for spectral class and spectral subclass classification; (3)χ² method based on nearest neighbor for luminosity typede termination. The experiments on low-resolution spectra show that the system achieves 3.2 spectral subclass precision for spectral and spectral subclass classification, 60% correct rate for luminosity recognition, and 78% rate for the luminosity recognition with error less than orequal to 1. The system is easy, rapid intraining, and feasible for thetomated spectra classification.