GuidedHybSensUIR Benchmark Dataset
收藏资源简介:
GuidedHybSensUIR Benchmark Dataset是由澳门大学和中国科学院深圳先进技术研究院等机构联合构建的水下图像恢复基准数据集。该数据集整合了来自四个真实世界水下图像数据集(UIEB、EUVP、LSUI和RUIE)的5600张训练图像、490对配对测试数据和460个无参考目标的测试样本,旨在为水下图像恢复任务提供标准化的基准。数据集的内容涵盖了多种水下环境和退化情况,能够有效支持深度学习模型的训练和评估。该数据集的构建过程包括从多个来源收集图像并进行配对处理,确保数据的多样性和代表性。该数据集的应用领域主要集中在水下图像恢复和增强,旨在解决水下图像因光线吸收和散射导致的颜色失真和模糊问题,提升水下视觉任务的准确性和效率。
The GuidedHybSensUIR Benchmark Dataset is an underwater image restoration benchmark dataset jointly constructed by the University of Macau, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences and other institutions. This dataset integrates 5600 training images, 490 paired test samples, and 460 reference-free test samples from four real-world underwater image datasets (UIEB, EUVP, LSUI, and RUIE), aiming to provide a standardized benchmark for underwater image restoration tasks. The dataset covers diverse underwater environments and degradation scenarios, which can effectively support the training and evaluation of deep learning models. The construction process of this dataset includes collecting images from multiple sources and performing paired processing, ensuring the diversity and representativeness of the data. The application scenarios of this dataset mainly focus on underwater image restoration and enhancement, aiming to address the color distortion and blur issues of underwater images caused by light absorption and scattering, and improve the accuracy and efficiency of underwater visual tasks.




