Evaluation Framework for Multiband Image Enhancement and Blending Algorithms in Enhanced Flight Vision Systems - Image Dataset

This dataset contains data used in the research published by MLabs Optronics in the paper: Medina Heierle, Victor, María Tejada Casado, Alberto Briasco González, Hugo Jestes Zoilo, Jesús Martín Tapia, Adeodato Altamirano Aguilar, and Javier Muñoz De Luna Clemente. Evaluation Framework for Multiband Image Enhancement and Blending Algorithms in Enhanced Flight Vision Systems. Proceedings of the 14th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), pp. 274-280. IEEE, 2018. The dataset is classified into 3 folders: - IR_VIS: Contains 28 pairs of images in the IR (some images may be in the NIR spectrum instead) and Visual spectrum, taken from different public repositories off the internet, which are typically used in multispectral fusion research. - Fusion: Contains 8 sets with the results of applying each of the 4 fusion algorithms described in the paper on some of the images in folder "IR_VIS". - VIS haze filtering: Contains 24 images taken with a CCD camera of a contrast target inside a fog simulation cabin in a laboratory. For comparison purposes, all images have been taken with a similar amount of fog, which is as much as was possible while still being able to see the target with the camera through the fog. Each image has been taken with a different type of filter (filter information is provided in another image inside the folder). Mlabs Optronics PTA Calle Pierre Laffitte, 8 29590 Málaga (Spain) www.mlabsoptronics.com [e-mail]

本数据集包含MLabs光电子（MLabs Optronics）发表的研究中所用的相关数据，相关论文信息如下：Medina Heierle, Victor、María Tejada Casado、Alberto Briasco González、Hugo Jestes Zoilo、Jesús Martín Tapia、Adeodato Altamirano Aguilar以及Javier Muñoz De Luna Clemente，论文题为《增强型飞行视觉系统中的多波段图像增强与融合算法评估框架》，收录于第14届国际信号-图像技术与基于互联网系统会议（SITIS）论文集，页码274-280，IEEE出版社，2018年。
本数据集分为3个文件夹：
- IR_VIS：包含28组红外（IR，部分图像可能属于近红外（NIR）光谱）与可见光光谱的图像对，这些图像均取自互联网公开资源库，是多光谱融合研究的常用数据集。
- Fusion：包含8组结果图像，对应将论文中介绍的4种融合算法应用于IR_VIS文件夹内部分图像后得到的输出结果。
- VIS haze filtering（可见光除雾滤波）：包含24张由CCD相机拍摄的图像，拍摄场景为实验室雾效模拟舱内的对比度测试靶标。为保证对比实验的一致性，所有图像的雾量均调整至当前相机可透过雾幕清晰识别靶标的最大可行水平。每张图像均使用了不同类型的滤光片，滤光片相关信息已存储在该文件夹内的另一张图像文件中。
MLabs Optronics PTA Calle Pierre Laffitte, 8 29590 西班牙马拉加
官网：www.mlabsoptronics.com
邮箱：[e-mail]