Dai Mingrui, Li Guohua, Shi Weifeng
Institute of Computing Technology, China Academy of Railway Sciences Co., Ltd., Beijing 100081, China.
Sensors (Basel). 2024 Sep 12;24(18):5930. doi: 10.3390/s24185930.
The density of fog is directly related to visibility and is one of the decision-making criteria for airport flight management and highway traffic management. Estimating fog density based on images and videos has been a popular research topic in recent years. However, the fog density estimated results based on images should be further evaluated and analyzed by combining weather information from other sensors. The data obtained by different sensors often need to be aligned in terms of time because of the difference in acquisition methods. In this paper, we propose a video and a visibility data alignment method based on temporal consistency for data alignment. After data alignment, the fog density estimation results based on images and videos can be analyzed, and the incorrect estimation results can be efficiently detected and corrected. The experimental results show that the new method effectively combines videos and visibility for fog density estimation.
雾的密度与能见度直接相关，是机场飞行管理和公路交通管理的决策标准之一。近年来，基于图像和视频估计雾密度一直是一个热门的研究课题。然而，基于图像的雾密度估计结果应结合其他传感器的天气信息进行进一步评估和分析。由于采集方式的不同，不同传感器获取的数据通常需要在时间上进行对齐。在本文中，我们提出了一种基于时间一致性的数据对齐方法，用于视频和能见度数据的对齐。数据对齐后，可以对基于图像和视频的雾密度估计结果进行分析，有效地检测和纠正错误的估计结果。实验结果表明，该新方法有效地结合了视频和能见度进行雾密度估计。
