The process of burn debridement is a challenging technique requiring significant skills to identify the regions that need excision and their appropriate excision depths. In order to assist surgeons, a machine learning tool is being developed to provide a quantitative assessment of burn-injured tissue. This paper presents three non-invasive optical imaging techniques capable of distinguishing four kinds of tissue—healthy skin, viable wound bed, shallow burn, and deep burn—during serial burn debridement in a porcine model. All combinations of these three techniques have been studied through a k-fold cross-validation method. In terms of global performance, the combination of all three techniques significantly improves the classification accuracy with respect to just one technique, from 0.42 up to more than 0.76. Furthermore, a non-linear spatial filtering based on the mode of a small neighborhood has been applied as a post-processing technique, in order to improve the performance of the classification. Using this technique, the global accuracy reaches a value close to 0.78 and, for some particular tissues and combination of techniques, the accuracy improves by 13%.

Pan-sharpening is a fundamental and crucial image processing task for many remote sensing applications, which generates a high-resolution multi-spectral image by fusing a low-resolution multi-spectral image and a high-resolution panchromatic image. Recently, vision transformers have been introduced into the pan-sharpening task for utilizing global contextual information. However, long-range and local dependencies modeling and multi-scale feature learning are all essential to the pansharpening task. Learning and exploiting these various information raises a big challenge and limits the performance and efficiency of existing pansharpening methods. To solve this issue, we propose a Pan-sharpening Network based on Multi-scale embedding and Dual attention transformers (MDPNet). Specifically, a multi-scale embedding block is proposed to embed multi-scale information of the images into vectors. Thus, transformers only need to process a multi-spectral embedding sequence and a panchromatic embedding sequence to efficiently use multi-scale information. Furthermore, an additive hybrid attention transformer is proposed to fuse the embedding sequences in an additive injection manner. Finally, a channel self-attention transformer is proposed to utilize channel correlations for high-quality detail generation. Experiments over QuickBird and WorldView-3 data sets demonstrate the proposed MDPNet outperforms state-of-the-art methods visually and quantitatively with low running time. Ablation studies further verify the effectiveness of the proposed multi-scale embedding and transformers in pan-sharpening.

Pansharpening aims to integrate low-resolution multispectral images (LRMS) and panchromatic images (PAN) to obtain the high-resolution multispectral images (HRMS). As PAN and LRMS images capture the same scene, they have some common information. Due to sensor characteristics differences, they also have unique information. Therefore, recent pansharpening methods separate common and unique features to reduce redundancy. However, they are not explicitly equipped with specific modules to extract common and unique features, which limits redundancy removal. To solve the problems, we propose a pansharpening method using a cross-attention-based module to specifically extract common and unique feature maps from the MS and PAN images. In addition, mutual information maximization and minimization constraints are used to enforce the common and unique features, which can reduce the redundant information. Finally, these features are concatenated to generate the HRMS. Extensive experimental results on QuickBird and Gaofen-2 datasets demonstrate that the proposed method outperforms other methods quantitatively and qualitatively.