Aiming at the physical truth of the variety condition of the cultural relics, gas sensor failure and qualitative imprecision of the corrosive gases, based on the Internet of Things technology and pattern recognition technology, in the case of considering the accuracy and the actual requirements of the project, a pattern recognition algorithm based on SAW sensor array is designed in this paper, Meanwhile, the simulation and verification are carried out based on MATLAB. It is shown that the best prediction results are obtained when the sensor array output is used as the actual input of the neural network, then the training and testing of this single layer BP neural network with six neurons in hidden layer are performed, The recognition rate of the corrosive gas is up to about 95%, which improves the monitoring level of corrosive gas in the field micro meteorological environment.