Infrared image enhancement through saliency feature analysis based on multi-scale decomposition

Fusion for visible and infrared images using visual weight analysis and bilateral filter-based multi scale decomposition

“…With frequency-tuned idea [14,23], one could obtain full resolution V map with well-defined boundaries of salient objects. From this theory, the largest salient object would be highlighted to form a V map .…”

“…However, the size of characteristic information in image is always small, this global idea would be out of action [14]. That is to say, we should highlight those high frequency information in our visual weight map by extracting small objects.…”

“…People have tried to use saliency preserving in multi-focus image fusion [13]. Especially, Zhao and his partners develop several visual saliency-based image enhancement [14,15] and image fusion [16,17] method. With saliency extraction and visual weight designing for image feature extraction, the fused result not only looks good, but also the objective evaluation is very good.…”

Infrared and Visible Image Fusion with a Generative Adversarial Network and a Residual Network

“…According to the fusion strategies and theories adopted [6], several representative infrared and visible image fusion algorithms have been proposed, including multi-scale transform- [7][8][9], sparse representation- [10,11], neural network- [12,13], subspace- [14,15], and saliency-based [16,17] methods, hybrid models [18,19], and other methods [20,21]. These methods are widely used and still studied by many researchers.…”

The optimal code searching method with an improved criterion of coded exposure for remote sensing image restoration

“…Image processing technologies have a wide application along with the process of image transmission, restoration, compression and reconstruction by various image devices [1,2]. For these image devices, the signal to noise ratio (SNR) index of captured image is in direct proportion to square root of exposure time.…”