Laser Scan Compression for Rail Inspection

Area of Science:

• Railway Engineering
• Data Compression
• Non-Destructive Testing

Background:

• Automated rail track inspection is vital for reducing maintenance costs and enhancing railway safety.
• Key challenges include efficient data acquisition, storage, and analysis from various track components.
• Laser triangulation scanners are essential for detailed rail profile measurement and defect detection.

Purpose of the Study:

• To introduce an effective data compression method specifically designed for laser triangulation scanner data of rail tracks.
• To enable storage of detailed rail track scans using standard image compression formats like PNG.
• To evaluate the compression ratios and impact on data quality for rail geometry and defect detection applications.

Main Methods:

• Developed a compression technique leveraging the inherent regularity of rail track data and sensor limitations (range, resolution).
• Transformed laser scanner data to be compatible with common image compression formats (e.g., PNG).
• Applied both lossy and lossless compression to rail geometry and defect detection datasets.

Main Results:

• Achieved a compression ratio of 7.5 for rail geometry computation scans, maintaining rail gauge reproducibility.
• Attained a compression ratio of 5.6 for defect detection scans without significant loss of visual quality.
• Lossless compression yielded ratios of 5.1 for geometry data and 3.8 for inspection data.

Conclusions:

• The proposed compression method offers a practical solution for managing large datasets from rail track inspections.
• Efficient data compression is feasible without compromising critical data for both geometric analysis and defect identification.
• This approach facilitates improved data handling and storage for automated railway maintenance systems.