Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Robust Subpixel Localization of Diagonal Markers in Large-Scale Navigation via Multi-Layer Screening and Adaptive Matching

Jing Tao, Banglei Guan, Yang Shang, Shunkun Liang, Qifeng Yu

TL;DR

The paper tackles robust diagonal marker localization in large-scale navigation where background interference and inefficient sliding-window matching hinder accuracy and speed. It introduces a three-tier framework comprising image preprocessing, multi-layer corner screening, and subpixel location, with gradient-domain filtering, dynamic illumination normalization, and structure-based feature condensation to enhance saliency, followed by an adaptive template generation and accelerated NCC-based refinement for subpixel precision. Key contributions include curvature-density based candidate screening, symmetry/congruence checks for geometric consistency, and a geometry- and photometry-driven adaptive template that reduces search space while preserving accuracy; subpixel offsets are obtained via a local quadratic extremum fit. Experiments on lab and UAV flight datasets demonstrate robust performance under clutter and illumination changes, achieving subpixel accuracy and centimeter-level pose estimates with improved computational efficiency relative to state-of-the-art methods, paving the way for reliable field-of-view measurement in large-scale navigation.

Abstract

This paper proposes a robust, high-precision positioning methodology to address localization failures arising from complex background interference in large-scale flight navigation and the computational inefficiency inherent in conventional sliding window matching techniques. The proposed methodology employs a three-tiered framework incorporating multi-layer corner screening and adaptive template matching. Firstly, dimensionality is reduced through illumination equalization and structural information extraction. A coarse-to-fine candidate selection strategy minimizes sliding window computational costs, enabling rapid estimation of the marker's position. Finally, adaptive templates are generated for candidate points, achieving subpixel precision through improved template matching with correlation coefficient extremum fitting. Experimental results demonstrate the method's effectiveness in extracting and localizing diagonal markers in complex, large-scale environments, making it ideal for field-of-view measurement in navigation tasks.

Robust Subpixel Localization of Diagonal Markers in Large-Scale Navigation via Multi-Layer Screening and Adaptive Matching

TL;DR

The paper tackles robust diagonal marker localization in large-scale navigation where background interference and inefficient sliding-window matching hinder accuracy and speed. It introduces a three-tier framework comprising image preprocessing, multi-layer corner screening, and subpixel location, with gradient-domain filtering, dynamic illumination normalization, and structure-based feature condensation to enhance saliency, followed by an adaptive template generation and accelerated NCC-based refinement for subpixel precision. Key contributions include curvature-density based candidate screening, symmetry/congruence checks for geometric consistency, and a geometry- and photometry-driven adaptive template that reduces search space while preserving accuracy; subpixel offsets are obtained via a local quadratic extremum fit. Experiments on lab and UAV flight datasets demonstrate robust performance under clutter and illumination changes, achieving subpixel accuracy and centimeter-level pose estimates with improved computational efficiency relative to state-of-the-art methods, paving the way for reliable field-of-view measurement in large-scale navigation.

Abstract

This paper proposes a robust, high-precision positioning methodology to address localization failures arising from complex background interference in large-scale flight navigation and the computational inefficiency inherent in conventional sliding window matching techniques. The proposed methodology employs a three-tiered framework incorporating multi-layer corner screening and adaptive template matching. Firstly, dimensionality is reduced through illumination equalization and structural information extraction. A coarse-to-fine candidate selection strategy minimizes sliding window computational costs, enabling rapid estimation of the marker's position. Finally, adaptive templates are generated for candidate points, achieving subpixel precision through improved template matching with correlation coefficient extremum fitting. Experimental results demonstrate the method's effectiveness in extracting and localizing diagonal markers in complex, large-scale environments, making it ideal for field-of-view measurement in navigation tasks.
Paper Structure (15 sections, 13 equations, 8 figures, 1 table)

This paper contains 15 sections, 13 equations, 8 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Method
  4. Image Preprocessing
  5. Multi-layer Point Screening
  6. Subpixel Location
  7. Template Adaptive Generation
  8. Subpixel Positioning Techniques
  9. Experiment
  10. Robustness Verification
  11. Precision Verification
  12. Checkerboard Positioning
  13. Flight Platforms Positioning
  14. Computational Efficiency
  15. Conclusion

Figures (8)

  • Figure 1: Flowchart of the automated diagonal marker extraction method in complex wide-area scenes.
  • Figure 2: Comparison of annular gradient scanning curves: (a) Grayscale curve of a pseudo-diagonal marker; (b) Grayscale curve of a diagonal marker.
  • Figure 3: Self-adaptive template generation diagram: (a) Determination of template parameters; (b) Template generation; (c) Correlation coefficient graph.
  • Figure 4: Imaging diagram of the actual project.
  • Figure 5: Stability test. (a) Different scenarios; (b) Different perspectives; (c) Complex environment.
  • ...and 3 more figures