Time-frequency fusion features-based incremental network for smartphone measured structural seismic response classification

•A novel time–frequency fusion features-based incremental network is proposed to classify structural seismic responses.•It can efficiently build the structure seismic response classification model.•It is more than 84 times in training speed and 2 times in inference speed than some well-known deep learning networks.•It has the feature of incremental learning, allowing the network structure to be updated as the dataset is updated.