Short-wave infrared and visible image fusion based on a dual-band polarization imaging sensor

Multi-modal sensors can provide more useful and detailed information. Visible images contain a large amount of detail and texture information. Short-wave infrared imaging has a strong penetrating ability and performs well in complex environments. Polarization detection can obtain polarization information which conventional imaging techniques cannot obtain. In this paper, we construct a dual-band polarization imaging sensor that can be used for visible light, short-wave infrared, and polarization imaging. An image registration algorithm is applied to obtain multi-modal images of the same scene. Multiple image fusion methods such as Laplacian pyramid transform (LPT), Intensity-Hue-Saturation (IHS) transform, principal component analysis (PCA), and pulse-coupled neural network (PCNN) are compared and analyzed with a qualitative assessment of subjective and objective metrics. Based on the experiments, the advantages and disadvantages of different fusion methods are discussed.