Route damage prediction system and method

• US 9,875,414 B2
• Filed: 04/15/2014
• Issued: 01/23/2018
• Est. Priority Date: 04/15/2014
• Status: Active Grant

• Current Assignee: GE Global Sourcing LLC (Westinghouse Air Brake Technologies Corporation)
• Original Assignee: General Electric Company
• Inventors: Naithani, Nidhi, Rao, Dattaraj Jagdish, Bind, Anju, Chaki, Sreyashi Dey
• Primary Examiner(s): Le, Vu
• Assistant Examiner(s): Mangialaschi, Tracy
• Application Number:
  US14/253,294
• Publication Number:
  US 20150294153A1
• Time in Patent Office: 1,379 Days
• Field of Search: None
• CPC Class Codes: B61L 23/042 Track changes detection
• CPC Class Codes: B61L 23/047 Track or rail movements
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/20221 Image fusion; Image merging
• CPC Class Codes: G06T 2207/30252 Vehicle exterior; Vicinity of vehicle
• CPC Class Codes: G06T 7/0008 checking presence/absence
• CPC Class Codes: G06T 7/55 from multiple images
• CPC Class Codes: G06V 20/588 Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road

Abstract

A route damage prediction system includes cameras, a conversion unit, and an analysis unit. The cameras obtain image data that include a route traveled upon by vehicles. The image data includes still images and/or video of the route obtained at different times. The conversion unit includes one or more computer processors configured to at least one of create wireframe model data or modify the image data into the wireframe model data representative of the route. The analysis unit includes one or more computer processors configured to examine changes in the wireframe model data to identify a historical trend of changes in the image data. The analysis unit is configured to compare the historical trend of the changes in the image data with designated patterns of changes in the wireframe model data to determine when to request at least one of repair, inspection, or maintenance of the route.

Claims

1. A system comprising:

one or more cameras configured to obtain image data within one or more fields of view of the one or more cameras that include a route that is traveled upon by plural different vehicles, the route including a track having one or more rails, the image data including at least one of still images or video of the route obtained at different times;

a conversion unit including one or more computer processors configured to at least one of create wireframe model data based on the image data or modify the image data into the wireframe model data representative of the route, the wireframe model data including substantially parallel line segments that represent the one or more rails in the image data, the conversion unit further configured to filter out a temporary external factor on the one or more rails in the image data at a location of the route, the temporary external factor filtered out by combining at least one of different sets of the image data of the location of the route acquired at the different times or different sets of the wireframe model data of the location of the route that are based on the sets of image data to produce combined wireframe model data, wherein the temporary external factor is other than damage to the location of the route and is absent from the one or more rails in at least one of the sets of the image data; and

an analysis unit including one or more computer processors configured to examine the line segments of the combined wireframe model data to identify changes to the line segments over time that represent a historical trend of changes to the one or more rails, wherein the analysis unit is configured to compare the historical trend of the changes to the one or more rails with designated patterns of changes in the wireframe model data to determine when to request at least one of repair, inspection, or maintenance of the route.

2. The system of claim 1, further comprising a communication unit configured to communicate a request signal to direct the at least one of repair, inspection, or maintenance of the route to be performed based on comparing the historical trend of the changes to the one or more rails with the designated patterns.

3. The system of claim 1, wherein the conversion unit is configured to at least one of create the wireframe model data or modify the image data into the wireframe model data by identifying at least one of pixels or other locations in the image data having image characteristics that are within designated ranges of each other, wherein the line segments correspond to groups of pixels or locations having image characteristics that are within the designated ranges of each other.

4. The system of claim 1, wherein the historical trend of changes to the one or more rails develops over time and includes at least one of changes in a number of the line segments representative of the one or more rails in the combined wireframe model data, changes in spacing between the line segments in the combined wireframe model data, changes in lengths of the line segments, changes in shapes of the line segments, or changes in gaps between the line segments.

5. The system of claim 1, wherein the designated patterns are predetermined patterns stored in a memory device, the analysis unit determining when to request at least one of repair, inspection, or maintenance of the route based on the historical trend of the changes to the one or more rails matching one of the predetermined patterns or more closely matching the one predetermined pattern over other predetermined patterns.

6. The system of claim 1, wherein the historical trend of changes to the one or more rails develops over time and includes at least one of changes in a number of the line segments representative of the one or more rails at a location of the route, decreases in lengths of the line segments representative of the one or more rails at the location, or changes in gaps between the line segments representative of the one or more rails at the location.

7. The system of claim 1, wherein the analysis unit is configured to compare the historical trend of the changes to the one or more rails to the designated patterns by comparing the changes to the line segments that represent the one or more rails to the designated patterns.

8. The system of claim 1, wherein the analysis unit is configured to identify the changes to the line segments based on one or more of the line segments at least one of bending, changing shape, or moving over time.

9. The system of claim 1, wherein the conversion unit is configured to combine the different sets of the image data of the location of the route to filter out the temporary external factor on the one or more rails by calculating an average or median image characteristic for at least some pixels in the different sets of the image data.

10. A method comprising:

receiving image data having one or more fields of view that include a route that is traveled upon by plural different vehicles, the route including a track having one or more rails, the image data including at least one of still images or video of the route obtained at different times;

at least one of creating wireframe model data based on the image data or modifying the image data into the wireframe model data representative of the route, the at least one of creating the wireframe model data or modifying the image data into the wireframe model data performed by one or more computer processors, the wireframe model data including substantially parallel line segments that represent the one or more rails in the image data;

filtering out a temporary external factor on the one or more rails in the image data at a location of the route to at least one of create the wireframe model data or modify the image data into the wireframe model data, the temporary external factor filtered out by combining at least one of different sets of the image data of the location of the route acquired at the different times or different sets of the wireframe model data of the location of the route that are based on the sets of image data to produce combined wireframe model data, wherein the temporary external factor is other than damage to the location of the route and is absent from the one or more rails in at least one of the sets of the image data;

examining the line segments of the combined wireframe model data to identify changes to the line segments over time that represent a historical trend of changes to the one or more rails; and

determining when to request at least one of repair, inspection, or maintenance of the route based on the historical trend of the changes to the one or more rails.

11. The method of claim 10, further comprising communicating a request signal to direct the at least one of repair, inspection, or maintenance of the route to be performed based on the historical trend of the changes to the one or more rails.

12. The method of claim 10, wherein the wireframe model data is created or the image data is modified into the wireframe model data by identifying at least one of pixels or other locations in the image data having image characteristics that are within designated ranges of each other and assigning a common image characteristic in the wireframe model data to the at least one of pixels or other locations having the image characteristics that are within the designated ranges of each other, wherein the at least one of pixels or locations having the common image characteristic form one of the line segments.

13. The method of claim 10, wherein the historical trend of changes to the one or more rails includes at least one of changes in a number of the line segments representative of the one or more rails in the combined wireframe model data, changes in spacing between the line segments in the combined wireframe model data, changes in lengths of the line segments, changes in shapes of the line segments, or changes in gaps between the line segments.

14. The method of claim 10, wherein the changes in the line segments of the combined wireframe model data include at least one of bending, changing shape, or moving of one or more of the line segments over time.

15. The method of claim 10, wherein the different sets of the image data of the location of the route are combined to filter out the temporary external factor on the one or more rails by calculating an average or median image characteristic for at least some pixels in the different sets of the image data.

16. A system comprising:

one or more computer processors configured to;

receive image data acquired at different times, the image data representing at least one of still images or video of a common section of a route traveled by vehicles, the route including a track having one or more rails;

create wireframe model data from the image data, the wireframe model data including substantially parallel line segments that represent the one or more rails in the image data;

examine the line segments of the wireframe model data to identify changes in the line segments of the wireframe model data over time, the changes including at least one of bending, changing shape, or moving of one or more of the line segments; and

examine the changes in the line segments of the wireframe model data to determine when to request at least one of repair, maintenance, or inspection of the common section of the route;

wherein the one or more computer processors are also configured to filter out a temporary external factor on the one or more rails in the image data at a location of the route, the temporary external factor filtered out by combining different sets of the image data of the location of the route acquired at the different times, wherein the temporary external factor is other than damage to the location of the route and is absent from the one or more rails in at least one of the sets of the image data, and wherein the wireframe model data that is examined to identify changes in the line segments includes the wireframe model data after filtering out the temporary external factor on the one or more rails.

17. The system of claim 16, wherein the one or more computer processors are also configured to examine the image data to identify pixels in the image data having image characteristics that are within a designated range of each other and to create the wireframe model data by assigning a first designated image characteristic to the pixels having the image characteristics that are within the designated range of each other and assigning a different, second designated image characteristic to the pixels having the image characteristics that are not within the designated range of each other, wherein the pixels having the first image characteristic correspond to one of the line segments.

18. The system of claim 16, wherein the one or more computer processors are also configured to compare the changes in the wireframe model data with designated changes associated with at least one of different types or different degrees of damage to the common section of the route.

19. The system of claim 16, wherein the one or more computer processors are configured to combine the different sets of the image data of the location of the route to filter out the temporary external factor on the one or more rails by calculating an average or median image characteristic for at least some pixels in the different sets of the image data.