Pixel-level image fusion

FusionGAN: A generative adversarial network for infrared and visible image fusion

Infrared and visible image fusion methods and applications: A survey

As a popular signal modeling technique, sparse representation (SR) has achieved great success in image fusion over the last few years with a number of effective algorithms being proposed. However, due to the patch-based manner applied in sparse coding, most existing SR-based fusion methods suffer from two drawbacks, namely, limited ability in detail preservation and high sensitivity to misregistration, while these two issues are of great concern in image fusion. In this letter, we introduce a recently emerged signal decomposition model known as convolutional sparse representation (CSR) into image fusion to address this problem, which is motivated by the observation that the CSR model can effectively overcome the above two drawbacks. We propose a CSR-based image fusion framework, in which each source image is decomposed into a base layer and a detail layer, for multifocus image fusion and multimodal image fusion. Experimental results demonstrate that the proposed fusion methods clearly outperform the SR-based methods in terms of both objective assessment and visual quality.

Image Fusion With Convolutional Sparse Representation

IFCNN: A general image fusion framework based on convolutional neural network

A new image fusion performance metric based on visual information fidelity

To assess the performance of image quality metrics (IQMs), some regressions, such as logistic regression and polynomial regression, are used to correlate objective ratings with subjective scores. However, some defects in optimality are shown in these regressions. In this correspondence, monotonic regression (MR) is found to be an effective correlation method in the performance assessment of IQMs. Both theoretical analysis and experimental results have proven that MR performs better than any other regression. We believe that MR could be an effective tool for performance assessment in the IQM research.

Monotonic Regression: A New Way for Correlating Subjective and Objective Ratings in Image Quality Research

The invention relates to an image fusion method based on vision similarity, belonging to the technical field of image processing. The image infusion method comprises the following steps of: carrying out initialized population process on original images to be fused; selecting 2n original images from the initialized population according to the vision similarity value of each image and pairing the 2n original images; exchanging pixels of sub-images aiming at each pair of the selected original images to obtain n pairs of new images, and placing the new images into a new population; discarding 2n images with the greatest vision similarity values from the new population according to the vision similarity value of each image in the new population to leave N images in the new population; repeating the steps for T times, and selecting the image with the least vision similarity as the fused image. The invention is simple and reliable, information that is absent in the source images can not be added in the fused image in the fusion process, and the effective information in the source images can be effectively and accurately extracted and fused into the needed image.

Yin Cao

Papers

A new image fusion performance metric based on visual information fidelity

Monotonic Regression: A New Way for Correlating Subjective and Objective Ratings in Image Quality Research

Image fusion method based on image similarity