3D track assessment method

Patent metadata

Field	Value
Publication number	US-10616557-B2
Application number	US-201916516686-A
Country	US
Kind code	B2
Filing date	Jul 19, 2019
Priority date	Feb 20, 2015
Publication date	Apr 7, 2020
Grant date	Apr 7, 2020

Abstract

A 3D track assessment method is disclosed for identifying and assessing features of a railway track bed based on 3D elevation and intensity data gathered from the railway track bed.

First claim

What is claimed is:

1. A method of detecting and defining planar regions along a railway track bed using a system for assessing a railway track bed, the method comprising the steps of:
a. inputting elevation data, rail head edge feature coordinates, and rail base edge feature coordinates to a processor wherein significant elevations due to the rail heads for each rail have been removed from the elevation data;
b. eliminating rail regions from the elevation data using the processor;
c. defining a rail base 2D sliding neighborhood window;
d. calculating a 3D gradient using the sliding neighborhood window using the processor to produce 3D gradient data using the processor;
e. calculating surface normals for the 3D gradient data using the processor; and
f. determining planar regions of a railway track bed as areas in the 3D gradient data that have 3D surface normal values greater than a surface normal threshold using the processor.

2. The method of claim 1 further comprising the steps of eliminating planar regions with a surface area less than a minimum area threshold using the processor; and eliminating planar regions with a surface area less than an isolated planar area threshold minimum area threshold using the processor.

3. The method of claim 1 further comprising the step of determining whether a planar region represents a wooden tie and, if so, eliminating tie plates regions.

4. The method of claim 1 further comprising the step of defining best fit clustered planar regions based on rectangular tie region models.

5. The method of claim 4 further comprising the step of defining tie bounding boxes based on best fit tie models and detected planar regions.

6. The method of claim 5 further comprising the step of calculating minimum least square fit planar tie surfaces for detected planar regions within each defined tie bounding box.

7. The method of claim 6 further comprising the step of calculating the physical parameters and locations of the defined tie bounding boxes.

8. A method of detecting railroad tie distress using a system for assessing a railway track bed, the method comprising the steps of:
a. inputting elevation data, longitudinal and transverse elevation map sample resolution data, rail base edge feature coordinates, detected tie bounding box data, and approximate tie surface plane data to a processor wherein significant elevations due to the rail heads for each rail have been removed from the elevation data;
b. comparing ties in the elevation data with a tie surface plane model by calculating the difference between the tie surface elevation data and the tie surface plane model using the processor;
c. identifying tie surface regions with elevations less than a tie surface plane minus a crack depth threshold as tie crack targets using the processor; and
d. determining physical parameters of crack targets using the processor.

9. The method of claim 8 further comprising the step of identifying tie surface regions with elevations greater a tie surface plane plus a ballast height threshold as ballast obscured areas using the processor.

10. The method of claim 8 further comprising the step of eliminating small area and small length tie crack targets using the processor.

11. The method of claim 8 further comprising the step of identifying tie end regions with crack target areas greater than a cracked area threshold using the processor.

12. The method of claim 8 further comprising the step of identifying tie surface regions between rails which deviate either above or below a planar surface approximation by an amount greater than a broken tie threshold using the processor.

13. The method of claim 8 further comprising the steps of:
e. determining whether ties in the elevation data are concrete ties; and
f. determining whether tie end surface normal angle differences in concrete ties in the elevation data are greater than a broken concrete tie surface normal angle threshold using the processor.

14. A method of detecting railway track bed features using a system for assessing a railway track bed, the method comprising the steps of:
a. inputting elevation data, longitudinal and transverse elevation map sample resolution data, rail base edge feature coordinates, detected tie bounding box data, and a 3D feature library to a processor wherein significant elevations due to the rail heads for each rail have been removed from the elevation data;
b. testing each of a plurality of railway track bed features in the 3D feature library against feature targets in the elevation data in sequence using the processor;
c. eliminating feature template matching scores for feature targets which are less than a 3D feature correlation threshold using the processor; and
d. calculating and storing physical parameters for a plurality of feature targets using the processor.

15. The method of claim 14 wherein the railway track bed features comprise features selected from the group consisting of railway track bed tie fasteners and railway track bed anchors.

16. The method of claim 14 further comprising the steps of determining whether a tie is a wooden tie or a concrete tie; setting the 3D feature library to wooden tie features if a determination is made that a wooden tie is present; and setting the 3D feature library to concrete tie features if a determination is made that a concrete tie is present.

Assignees

• Tetra Tech

Inventors

• Mesher Darel

Classifications

• G06T7/0004

  Industrial image inspection · CPC title

• H04N13/189

  Recording image signals; Reproducing recorded image signals · CPC title

• G06T2207/30136

  Metal · CPC title

• G06T2207/10028

  Range image; Depth image; 3D point clouds · CPC title

• B61K9/08 Primary

  Measuring installations for surveying permanent way · CPC title