Fishing Trajectories (AIS)
A Pre-Labelled Dataset for Semantic Segmentation
By [source]
About this dataset
This dataset contains 128 labeled fishing trajectories that have been meticulously and closely examined for comprehensive extracts of detailed data about fishing activities. By manual labeling, our team is able to understand, interpret, and observe any changes happening in the marine environment.
The dataset includes id, timestep (t), longitude, latitude, x and y coordinates, signed turn that indicate turns taken by the boat’s trajectory since previous points of a segmented trajectory are stitched together, bearing created from Cartesian coordinates which indicates direction in which the boat’s trajectory is leading to or from a point from origin., time gap between two points measured in elapsed seconds or milliseconds with respect to origin point on the same segment , distance gap between two points measured in sea stones with respect to origin point on same segment , Euclidean speed as opposed to normal speed measured per second towards next coordinate on a constructed line considered straight by mathematicians , Last but not least; Distance-to-shore - described as arc connecting one computer location on land specified marine environment with calculate path or set of arcs that build shortest distance connection called geodesic between two locations. labels corresponding timestamp (t) – i.e 'fishing', 'not_fishing', etc . The combination of this vast data offers user precise information needed for accurate AI models attempted at prescience observed behaviors & valuable insights associated with them
More Datasets
For more datasets, click here.
Featured Notebooks
- 🚨 Your notebook can be here! 🚨!
How to use the dataset
This dataset contains labeled fishing trajectories that have been collected from AIS (Automatic Identification System) data. The data can be used to analyze and interpret fishing activities.
Research Ideas
- Unsupervised Anomaly Detection: By providing the labeled trajectories, machine learning models can detect strange behaviors in fishing activities which may indicate suspicious human activities and target them with flagging and fines.
- Prediction of Fishing Locations: Through analysis of the labeled data, machine learning models can predict probable fishing locations based on characteristics such as time gaps between points, euclidean speed, and distance to shore. This can be used by Governmental Agencies in order to better monitor fisheries and detect any illegal activity taking place there.
- Semantic Segmentation of Fishing Activities: The given labels for each trajectory such as trolling or drifting can be used to prepare a dataset for semantic segmentation in order to classify a type of behavior within a given trajectory for further investigation purposes when combined with GPS locations or other data sources about vessel performance or marine environment conditions
Acknowledgements
If you use this dataset in your research, please credit the original authors.
Data Source
License
License: CC0 1.0 Universal (CC0 1.0) - Public Domain Dedication
No Copyright - You can copy, modify, distribute and perform the work, even for commercial purposes, all without asking permission. See Other Information.
Columns
File: 128_fishing_trajs.csv
| Column name |
Description |
| t |
Timestep (Integer) |
| longitude |
Longitude coordinate of the point (Float) |
| latitude |
Latitude coordinate of the point (Float) |
| x |
X coordinate of the point (Float) |
| y |
Y coordinate of the point (Float) |
| signed_turn |
Signed turn of the trajectory (Integer) |
| bearing |
Bearing of the trajectory (Integer) |
| time_gap |
Time gap between points (Float) |
| distance_gap |
Distance gap between points (Float) |
| euc_speed |
Euclidean speed of the trajectory (Float) |
| distanceToShore |
Distance to shore of the point (Float) |
| label |
Label of the trajectory (String) |
Acknowledgements
If you use this dataset in your research, please credit the original authors.
If you use this dataset in your research, please credit .