Deep transformer and few‐shot learning for hyperspectral image classification

Recently,deep learning has achieved considerable results in the hyperspectral image(HSI)classification.However,most available deep networks require ample and authentic samples to better train the models,which is expensive and inefficient in practical tasks.Existing few‐shot learning(FSL)methods generally ignore the potential relationships between non‐local spatial samples that would better represent the underlying features of HSI.To solve the above issues,a novel deep transformer and few‐shot learning(DTFSL)classification framework is proposed,attempting to realize fine‐grained classification of HSI with only a few‐shot instances.Specifically,the spatial attention and spectral query modules are introduced to overcome the constraint of the convolution kernel and consider the information between long‐distance location(non‐local)samples to reduce the uncertainty of classes.Next,the network is trained with episodes and task‐based learning strategies to learn a metric space,which can continuously enhance its modelling capability.Furthermore,the developed approach combines the advantages of domain adaptation to reduce the variation in inter‐domain distribution and realize distribution alignment.On three publicly available HSI data,extensive experiments have indicated that the proposed DT‐FSL yields better results concerning state‐of‐the‐art algorithms.

Wave-particle duality relation with a quantum N-path beamsplitter

The wave-particle duality relation derived by Englert sets an upper bound of the extractable information from wave and particle properties in a two-path interferometer.Surprisingly,previous studies demonstrated that the introduction of a quantum beamsplitter in the interferometer could break the limitation of this upper bound,due to interference between wave and particle states.Along the other line,a lot of efforts have been made to generalize this relation from the two-path setup to the N-path case.Thus,it is an interesting question that whether a quantum N-path beamsplitter can break the limitation as well.This paper systemically studies the model of a quantum N-path beamsplitter,and finds that the generalized wave-particle duality relation between interference visibility and path distinguishability is also broken in certain situations.We further study the maximal extractable information's reliance on the interference between wave and particle properties,and derive a quantitative description.We then propose an experimental methodology to verify the break of the limitation.Our work reflects the effect of quantum superposition on wave-particle duality,and exhibits a new aspect of the relation between visibility and path distinguishability in N-path interference.Moreover,it implies the observer's influence on wave-particle duality.