five

Labels for wave-scarp interaction images

收藏
Mendeley Data2024-01-31 更新2024-06-29 收录
下载链接:
https://figshare.com/articles/Labels_for_wave-scarp_interaction_images/12765494/1
下载链接
链接失效反馈
官方服务:
资源简介:
Beach and surf cameras offer a way to visualize interactions between waves and scarps (both dune and beach scarps). This dataset is a single csv file that lists 1) publicly available frames of video from Buxton, North Carolina, USA; 2) a human coded label describing whether ocean waves are interacting with a beach or dune scarp. A total of 984 frames have been labeled The data table has 6 columns: 1) the name of the catalog (Buxton Coastal Camera); 2) the archive name, which corresponds to the specific video that supplies the frames used for coding. The videos are demarcated using year, month, day, time format, and correspond to 10 minute increments (which is the time interval provided during data retrieval); 3) the frame number from that video segment; Columns 4 through 6 indicate the coding of the image based on wave-scarp interaction seen in the frame. There are three options: ‘NoInteraction’ (waves are not impacting the scarp), Interaction (waves are impacting the scarp), or ‘Close’ (the waves are close to the scarp, and judged by the human labeler to have a high likelihood of interacting with the scarp in subsequent frames). The publicly available video from Buxton, North Carolina, USA is from the NOAA NOS (National Ocean Service) and SECOORA (The Southeast Coastal Ocean Observing Regional Association) camera monitoring system — WebCAT (Web Camera Applications Testbed). The link to the WebCAT site is found below, as well as a link to a programmatic tool (PyWebCAT) for retrieving video and slicing video into frames. We encourage users of this dataset to use PyWebCAT to download video, extract frames, and then merge with this dataset. Hand labeling was done using the Coastal Image Labeler (link below). We thank NOAA NOS, SECOORA and the WebCAT project for making camera footage publicly available and supporting open science and data.
创建时间:
2024-01-31
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作