Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ALINA: Advanced Line Identification and Notation Algorithm

Mohammed Abdul Hafeez Khan, Parth Ganeriwala, Siddhartha Bhattacharyya, Natasha Neogi, Raja Muthalagu

TL;DR

A novel annotation framework, Advanced Line Identification and Notation Algorithm (ALINA), is proposed, which can be used for labeling taxiway datasets that consist of different camera perspectives and variable weather attributes (sunny and cloudy).

Abstract

Labels are the cornerstone of supervised machine learning algorithms. Most visual recognition methods are fully supervised, using bounding boxes or pixel-wise segmentations for object localization. Traditional labeling methods, such as crowd-sourcing, are prohibitive due to cost, data privacy, amount of time, and potential errors on large datasets. To address these issues, we propose a novel annotation framework, Advanced Line Identification and Notation Algorithm (ALINA), which can be used for labeling taxiway datasets that consist of different camera perspectives and variable weather attributes (sunny and cloudy). Additionally, the CIRCular threshoLd pixEl Discovery And Traversal (CIRCLEDAT) algorithm has been proposed, which is an integral step in determining the pixels corresponding to taxiway line markings. Once the pixels are identified, ALINA generates corresponding pixel coordinate annotations on the frame. Using this approach, 60,249 frames from the taxiway dataset, AssistTaxi have been labeled. To evaluate the performance, a context-based edge map (CBEM) set was generated manually based on edge features and connectivity. The detection rate after testing the annotated labels with the CBEM set was recorded as 98.45%, attesting its dependability and effectiveness.

ALINA: Advanced Line Identification and Notation Algorithm

TL;DR

A novel annotation framework, Advanced Line Identification and Notation Algorithm (ALINA), is proposed, which can be used for labeling taxiway datasets that consist of different camera perspectives and variable weather attributes (sunny and cloudy).

Abstract

Labels are the cornerstone of supervised machine learning algorithms. Most visual recognition methods are fully supervised, using bounding boxes or pixel-wise segmentations for object localization. Traditional labeling methods, such as crowd-sourcing, are prohibitive due to cost, data privacy, amount of time, and potential errors on large datasets. To address these issues, we propose a novel annotation framework, Advanced Line Identification and Notation Algorithm (ALINA), which can be used for labeling taxiway datasets that consist of different camera perspectives and variable weather attributes (sunny and cloudy). Additionally, the CIRCular threshoLd pixEl Discovery And Traversal (CIRCLEDAT) algorithm has been proposed, which is an integral step in determining the pixels corresponding to taxiway line markings. Once the pixels are identified, ALINA generates corresponding pixel coordinate annotations on the frame. Using this approach, 60,249 frames from the taxiway dataset, AssistTaxi have been labeled. To evaluate the performance, a context-based edge map (CBEM) set was generated manually based on edge features and connectivity. The detection rate after testing the annotated labels with the CBEM set was recorded as 98.45%, attesting its dependability and effectiveness.
Paper Structure (19 sections, 12 equations, 5 figures, 4 tables, 1 algorithm)

This paper contains 19 sections, 12 equations, 5 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Methodology
  4. Frame Representation
  5. Interactive ROI Definition on the Initial Frame
  6. Perspective Transformation of the ROI
  7. Color Feature Normalization
  8. HSV-based Color Thresholding
  9. Histogram-based Analysis of the Thresholded ROI
  10. Identifying Line Markings and Mitigating False Detections
  11. CIRCLEDAT: Circular Threshold Pixel Discovery and Traversal Algorithm
  12. Frame Unwarping and Annotating the Taxiway Line Marking Pixels
  13. Experimental Results
  14. ALINA Performance: Specifications and Scenarios
  15. Ablation Study on Threshold Value
  16. ...and 4 more sections

Figures (5)

  • Figure 1: (a) Illustrates the aircraft used for this research, (b) and (c) displays the setup of cameras at different perspectives inside the aircraft, (d) shows the manually defined region of interest on a frame, (e) showcases the annotation label created by ALINA on a frame.
  • Figure 2: Framework of ALINA
  • Figure 3: (a),(e),(i): ROI Definition, (b),(f),(j): Perspective Transformation, (c),(g),(k): Color Feature Normalization, (d),(h),(l): HSV-based Color Thresholding
  • Figure 4: (a),(d),(g): CIRCLEDAT, (b),(e),(h): Frame Unwarping, (c),(f),(i): Taxiway Line Marking Annotation
  • Figure 5: (a) Manually Outlined Contour Region (b) CBEM