用于评估高光谱图像中血液检测的数据集
收藏资源简介:
本数据集由理论与应用信息学研究所,波兰科学院创建,旨在为高光谱图像中的血液检测算法开发提供支持。数据集包含多个检测场景,涵盖不同复杂度,允许测试机器学习方法在不同采集环境、背景类型、血液年龄及类似血液物质存在时的表现。数据集通过在线开放许可提供,包含14个高光谱图像,每个约180MB,以及实验源代码和地面实况图,以便复现论文中的结果。该数据集的应用领域包括法医学中的血液检测,旨在解决血液检测中的复杂性和高维度问题,提高检测算法的准确性和可靠性。
This dataset was created by the Institute of Theoretical and Applied Informatics, Polish Academy of Sciences, to support the development of blood detection algorithms for hyperspectral imagery. It encompasses multiple detection scenarios with varying levels of complexity, enabling the assessment of machine learning models' performance across diverse acquisition environments, background types, blood aging stages, and scenarios involving blood-like substances. The dataset is distributed under an open online license, and consists of 14 hyperspectral images (each approximately 180 MB in size), alongside experimental source code and ground truth maps to facilitate the reproduction of results presented in peer-reviewed papers. Its application domains include blood detection in forensic science, where it aims to resolve the complexity and high-dimensionality challenges inherent in blood detection tasks, thereby improving the accuracy and reliability of detection algorithms.




