Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A horizon line annotation tool for streamlining autonomous sea navigation experiments

Yassir Zardoua, Abdelhamid El Wahabi, Mohammed Boulaala, Abdelali Astito

TL;DR

The paper tackles robust horizon-line detection for autonomous maritime navigation, where datasets under diverse sea conditions are scarce and annotations are error-prone. It presents Horizon Annotator, a GUI-based tool organized into load, display, annotation, and browse blocks, featuring a comprehensive keymap and a workflow for rapid frame-by-frame labeling. Annotation data are saved in $LineGT.npy$ with per-frame parameters $Y$, $φ$, $x_s$, $y_s$, $x_e$, $y_e$, using $np.nan$ to denote unannotated frames. This tool aims to accelerate the development of reliable horizon detectors by enabling fast, accurate labeling across challenging sea states and promoting reproducible research through publicly available tooling.

Abstract

Horizon line (or sea line) detection (HLD) is a critical component in multiple marine autonomous navigation tasks, such as identifying the navigation area (i.e., the sea), obstacle detection and geo-localization, and digital video stabilization. A recent survey highlighted several weaknesses of such detectors, particularly on sea conditions lacking from the most extensive dataset currently used by HLD researchers. Experimental validation of more robust HLDs involves collecting an extensive set of these lacking sea conditions and annotating each collected image with the correct position and orientation of the horizon line. The annotation task is daunting without a proper tool. Therefore, we present the first public annotation software with tailored features to make the sea line annotation process fast and easy. The software is available at: https://drive.google.com/drive/folders/1c0ZmvYDckuQCPIWfh_70P7E1A_DWlIvF?usp=sharing

A horizon line annotation tool for streamlining autonomous sea navigation experiments

TL;DR

The paper tackles robust horizon-line detection for autonomous maritime navigation, where datasets under diverse sea conditions are scarce and annotations are error-prone. It presents Horizon Annotator, a GUI-based tool organized into load, display, annotation, and browse blocks, featuring a comprehensive keymap and a workflow for rapid frame-by-frame labeling. Annotation data are saved in with per-frame parameters , , , , , , using to denote unannotated frames. This tool aims to accelerate the development of reliable horizon detectors by enabling fast, accurate labeling across challenging sea states and promoting reproducible research through publicly available tooling.

Abstract

Horizon line (or sea line) detection (HLD) is a critical component in multiple marine autonomous navigation tasks, such as identifying the navigation area (i.e., the sea), obstacle detection and geo-localization, and digital video stabilization. A recent survey highlighted several weaknesses of such detectors, particularly on sea conditions lacking from the most extensive dataset currently used by HLD researchers. Experimental validation of more robust HLDs involves collecting an extensive set of these lacking sea conditions and annotating each collected image with the correct position and orientation of the horizon line. The annotation task is daunting without a proper tool. Therefore, we present the first public annotation software with tailored features to make the sea line annotation process fast and easy. The software is available at: https://drive.google.com/drive/folders/1c0ZmvYDckuQCPIWfh_70P7E1A_DWlIvF?usp=sharing
Paper Structure (13 sections, 10 figures, 1 table)

This paper contains 13 sections, 10 figures, 1 table.

Table of Contents

  1. Introduction
  2. Definition of the horizon line
  3. Applications of the horizon line
  4. The need of custom horizon annotation tool
  5. Description of the application: interface blocks and features
  6. Load directories block
  7. Image display block
  8. Annotation block
  9. Browse block
  10. Annotation files
  11. Keymap for quick and easy annotation
  12. Message boxes
  13. Conclusion

Figures (10)

  • Figure 1: Definition and representation of the horizon: the horizon line separates the sea from the sky (a) and the sea from the coast (b); (c) $\{Y, \phi\}$ representation on Cartesian coordinates $xy$ of images
  • Figure 2: The GUI before video loading
  • Figure 3: The Load directories block
  • Figure 4: The GUI after loading (a) a video file and an (b) existing annotation file.
  • Figure 5: The process of line drawing: (a) before releasing the left mouse button; (b) after releasing the left mouse button; (c) the mouse position coordinates.
  • ...and 5 more figures