Using infrared polarization information(degree of polarization, polarization angle)to image a target can better suppress the background noise of the image and improve the signal-to-noise ratio. In addition, the polarization information generally contains richer edge contour information than the light intensity information. Therefore, an algorithm is proposed to fuse the infrared radiation intensity image and polarization image. This method first performs Laplacian pyramid decomposition on each image participating in the fusion to obtain a decomposed image of each layer; then uses different fusion methods for image fusion for each decomposed image to obtain a fused image of each layer.Image reconstruction is performed on the fused image of each layer to get the final fusion result. The effect of multiple image fusion shows that this method can increase the amount of information in the image, which is beneficial to scene perception and target recognition.