Rail feature identification system

1. A rail vehicle comprising:

a frame coupled to a plurality of rail wheels configured to move along a rail track; and

a railroad track feature detection system coupled to the frame, the feature detection system comprising:

a camera coupled to the frame and positioned such that the camera is configured to capture a plurality of images of a rail track as the rail vehicle moves along the rail track; and

a computer having: at least one memory comprising instructions; and at least one processor configured to execute the instructions, wherein the instructions cause the at least one processor to:

receive the plurality of images, and for each of the images:

assign a location identifier;

detect one or more rail features;

identify the one or more rail features as a relevant rail feature from a dataset of predetermined rail features; and

assign a feature identifier to each identified rail feature;

based on the location identifier and the feature identifier:

determine a geographical location of the one or rail more features; and

track the one or more rail features;

assign a first location identifier to a first image of the plurality of images;

identify, in the first image, the one or more rail features;

assign a second location identifier to a second image of the plurality of images;

compare the first location identifier with the second location identifiter to determine a first displacement distance between the identified rail features in the first image;

based on the first displacement distance, determine if one or more rail features in the second image correspond to one of the one or more identified rail features in the first image; and

based on the determintion that:

a first subset of the one or more rail features in the second image correspond to at least one of the one or more identified rail features in the first image, setting the first subset as detected rail features and re-identifying the first subset of the one or more rail features; and

a second subset of the one or more features in the second image does not correspond to at least one of the one or more identified rail features in the first image, setting the second subset as detected rail features and identifying, in the first image, the second subset of the one or more rail features.

2. The rail vehicle of claim 1 , wherein the instructions include a convolutional neural network.

3. The rail vehicle of claim 1 , wherein:

the processor is configured to identify the one or more rail features in real-time; and

for the first image of the plurality of images, the processor is configured to:

classify a first set of one or more ojbects in the first image as one of the one or more rail features;

classify a second set of the one or more objects in the first image as extraneous features; and

filter out the extraneous features.

4. The rail vehicle of claim 1 , wherein the processor is configured to:

determine if one of the rail features is missing or displaced at a first rail plate; and

based on the determination that one of the rail features is missing or displaced at the first rail plate, generate an alert signal.

5. The rail vehicle of claim 4 , further comprising:

a work unit coupled to the frame and configured to perform one or more railroad maintenance or operation functions; and

wherein based on the alert signal, the processor is configured to control the work unit to perform a maintenance function at the first rail plate.

6. The rail vehicle of claim 1 , further comprising:

a second camera;

a third light source; and

a fourth light source; and

wherein:

the frame includes a second bracket assembly;

the third light source and fourth light source disposed on the second bracket assembly such that light emitted from the third and fourth light source is directed to opposing sides of a second rail of the rail track; and

the second camera is disposed between the third light source and the fourth light source and configured to take images of the second rail.

7. A railroad track feature detection system comprising:

an imaging device configured to be coupled to a frame of a rail vehicle and positioned such that the imaging device is configured to capture a plurality of images of a rail track as the rail vehicle moves along the rail track; and

a computer having:

at least one memory comprising instructions, at least a portion of the instructions configured as a neural network; and

at least one processor configured to execute the instructions, wherein the instructions cause the at least one processor to perform one or more operations in real-time;

wherein the one or more operations include:

receiving the plurality of images, and for each of the images:

assigning a location identifier; and

identifying one or more rail features that correspond to one of a plurality of predetermined rail features; and

determining a location of each of the one or more identified rail features;

assigning a first location identifier to a first image of the plurality of images,

identifying, in the first image, the one or more rail features;

assigning a second location identifier to a second image of the plurality of images,

comparing the first location identifier with the second location identifier to determine a first displacement distance between the identified rail features in the first image;

based on the first displacement distance, determining if one or more rail features in the second image correspond to one of the one or more identified rail features in the first image;

based on the determintion that:

a first subset of the one or more rail features in the second image correspond to at least one of the one or more identified rail features in the first image, setting the first subset as detected rail features and re-identifying the first subset of the one or more rail features; and

a second subset of the one or more rail features in the second image does not correspond to at least one of the one or more identified rail features in the first image, setting the second subset as detected rail features and identifying, in the first image, the second subset of the one or more rail features.

8. The system of claim 7 , further comprising:

assigning a third location identifier to a third image of the plurality of images,

comparing the third location identifier to the first and second location identifiers to determine a second displacement distance;

based on the second displacement distance, determining if the one or more rail features in the third image corresponds to one of the rail features of the first and second subsets; and

based on the determination that:

the first subset of the one or more rail features is not contained within the third image, setting the first subset of rail features as tracked rail features;

a third subset of the one or more rail features in the third image corresponds to at least one of the one or more identified rail features in the first image or the second image, re-identifying the third subset of the one or more rail features; and

a fourth subset of the one or more rail features in the third image does not correspond to at least one of the one or more identified rail features in the first and second images, identifying, in the third image, the fourth subset of the one or more rail features.

9. The system of claim 7 , wherein identifying one or more rail features includes identifying a rail plate, a hole defined by the rail plate, a colored marker disposed on a portion of the rail, an anchor, a work tie, and/or a joint bar.

10. The system of claim 7 , wherein the computer includes an object detector module, a color detector module, and a tracker module.

11. The system of claim 10 , wherein the object detector module is configured to:

receive the plurality of images;

detect, in real-time, one or more objects in a latest image of the plurality of images;

classify a first set of the one or more objects as one of the one or more rail features;

classify a second set of the one or more objects as extraneous features; and

output the one or more rail features.

12. The system of claim 11 , wherein the object detector module is configured as a convolutional neural network.

13. A railroad track feature detection system comprising:

an imaging device configured to be coupled to a frame of a rail vehicle and postioned such that the imaging device is configured to capture a plurality of images of a rail track as the rail vehicle moves along the rail track; and

a computer having:

at least one memory comprising instuctions, at least a portion of the instructions configured as a neural network; and

at least one processor configured to execute the instructions, wherein the instructions cause the at least one processor to perform one or more operation in real-time;

wherein the one or more operations include:

receiving the plurality of images, and for each of the images:

assigning a location identifier; and

identifying one or more rail features that correspond to one of a plurality of predetermined rail features; and

determining a location of each of the one or more identified rail features wherein the computer includes an object detector module, a color detector module, and a tracker module; and

wherein the color detector module is configured to:

receive the plurality of images;

determine a hue/saturation/value (HSV) of a plurality of pixels within a latest image of the plurality of images;

detect, in real-time, one or more color markers in the latest image; and

output the one or more color markers.

14. The system of claim 13 , wherein the tracker module is configured to:

receive the one or more rail features from the object detector module;

receive the one or more color markers from the color detector module;

assign a feature ID to each of the one or more rail features and the one or more color markers; and

track a location of each of the one or more rail features and the one or more color markers.

15. The system of claim 14 , wherein the tracker module is configured to:

generate a map of the rail track including one or more rail features, based on the location of each of the one or more rail features and the one or more color markers; and

transmit the map to an external source.

16. The system of claim 14 , wherein the feature ID includes location information, a feature identifier that corresponds to a predetermined rail feature, and a confidence score.

17. The system of claim 14 , wherein for each image of the plurality of images, the tracker module is configured to:

record the feature IDs associated with the one or more rail features and the one or more color markers within the image;

update an output list to include the feature IDs of a most recent image; and

annotate the image to include:

a frame around each of the one or more rail features and the one or more color markers within the image;

a ID number associated with the feature ID of the one or more rail features and the one or more color markers within the image; and

a confidence score associated with the feature ID of the one or more rail features and the one or more color markers within the image.

18. The system of claim 17 , wherein the tracker module is configured to transmit the annotated image to an external device.

19. A method of tracking one or more rail features, the method comprising:

receiving a first image from a camera module;

detecting one or more first rail features in the first image;

identifying the one or more first rail features;

receiving a second image from the camera module;

calculating a displacement distance associated with the first and second images;

detecting one or more second rail features in the second image;

based on the displacement distance, identifying a first subset of the one or more second rail features that correspond to one of the one or more first rail features;

re-identifying the first subset of the one or more second rail features; and

identifying a second subset of the one or more second rail features that do not correspond to one of the one or more first rail features;

identifying a first subset of the one or more first rail features that are not included in the second image; and

setting the first subset of the one or more first rail features as tracked feature and transmitting feature data of the tracked features to a display unit.