Technique for image fusion between gray-scale visual light and infrared images based on NSST and improved RF

Abstract

A new technique for image fusion between gray-scale visual light and infrared images based on non-subsampled shearlet transform (NSST) and improved receptive field (RF) is proposed in this paper. As a novel multi-scale geometric analysis tool, NSST can be optimally efficient in representing images containing edges and capturing the geometric features of multidimensional data. As a result, NSST is introduced into the field of image fusion in this paper and utilized to obtain the fused image. The source images could be decomposed in any scale and any direction, so the detailed information can be caught easily. Then, the basic traditional RF model is improved to be improved RF (IRF) which has much fewer parameters and a more effective structure compared with RF. Thirdly, with IRF and the model of local directional contrast (LDC), the fused sub-images can be achieved. Finally, the final fused image can be obtained by using inverse NSST to all fused sub-images. The numerical experiments demonstrate that the new technique presented in this paper is very competitive in image fusion applications in terms of both fusion performance and computational efficiency.