ELAS数据集

ELAS数据集是由巴西联邦大学圣埃斯皮里图分校的研究团队开发，旨在支持其提出的实时自我车道分析系统（ELAS）的研究。该数据集包含超过15,000帧的图像，涵盖多种驾驶场景，如城市道路、高速公路、交通、阴影等。数据集中的图像经过手动标注，用于评估车道估计、车道变化、车道中心化、路面标记、交叉口、车道标记类型和相邻车道等事件。此数据集的创建旨在填补文献中车道数据集的不足，并促进相关研究社区的公平比较。ELAS数据集的应用领域主要集中在自动驾驶和高级驾驶辅助系统（ADAS）中，旨在解决车道检测、车道保持和车道变化检测等关键问题。

The ELAS dataset was developed by a research team from the Federal University of Espírito Santo, Brazil, to support the research of their proposed real-time self-lane analysis system (ELAS). This dataset contains over 15,000 image frames covering a wide range of driving scenarios, such as urban roads, highways, traffic conditions, shadows, and others. All images in the dataset are manually annotated, which are used to evaluate tasks and events including lane estimation, lane changing, lane centering, road markings, intersections, lane marker types, and adjacent lanes. The creation of this dataset aims to fill the gap of existing lane-related datasets in academic literature and promote fair comparisons across the relevant research community. The ELAS dataset is mainly applied in autonomous driving and Advanced Driver Assistance Systems (ADAS), with the goal of addressing key challenges such as lane detection, lane keeping, and lane change detection.