Spatial Infrared Objects Discrimination based on Multi-Channel CNN with Attention Mechanism

•A new spatial infrared objects classification framework based on multi-channel CNN with attention mechanism is proposed.•The multi-band infrared radiation intensity sequence model of spatial object is established.•The proposed method has more higher recognition accuracy than other classical methods. Moreover, the proposed method has more stable recognition performance for noise interference, data loss and band mismatch.•The significance of multi-channel design and attention mechanism for the classification of multi-band infrared radiation intensity sequences is demonstrated.