Technique for image fusion based on NSST domain improved fast non-classical RF

• NSST is an excellent tool for multi-scale geometric analysis of images.
• Non-classic RF is utilized to improve the basic RF model.
• The NSST-non-classic-RF-based algorithm is devised and given in detail.
• The NSST-non-classic-RF model is more effective compared with other classic ones.

In this paper, a novel technique for image fusion based on NSST (Non-Subsampled Shearlet Transform) domain improved fast non-classical RF (Receptive Field) is proposed to resolve the problem of the multi-sensor image fusion. 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 final fused image. Multi-scale and multi-directional sparse decompositions of source images are performed by NSST to obtain the main and detailed information. Then, the basic traditional RF model is modified to be an improved fast non-classical RF (IFRF) which has much fewer parameters and a more effective structure compared with RF. Thirdly, with IFRF and the model of local directional contrast (LDC), the fused sub-images can be achieved. Meanwhile, the algorithm for image fusion based on NSST domain improved fast non-classical RF is devised. 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 has not only much better performance, but good running efficiency.