The application of fuzzy pattern recognition to ultrasonic detection for bonding quality of thin composite plate is studied. In this paper, firstly, the fuzzy membership function between each characteristic and bonding quality is established by BP neural network. Simulation results show that this method is convenient, simple, and it conforms to the practical application. Accordingly，the fuzzy subsets of standard modules and unknown bonding quality modules are established. Secondly, the fuzzy pattern recognition algorithm, which is designed by the nearest principle as the judging standard to judge bonding quality, is given. Experimental results show that the algorithm is very exact for quantitative recognition of bonding quality.