Reliability in detecting rail crossing events

• US 8,693,725 B2
• Filed: 04/19/2011
• Issued: 04/08/2014
• Est. Priority Date: 04/19/2011
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: Bobbitt, Russell P., Kjeldsen, Frederik C. M., Zhai, Yun
• Primary Examiner(s): Lu, Tom Y
• Application Number:
  US13/089,597
• Publication Number:
  US 20120269383A1
• Time in Patent Office: 1,085 Days
• Field of Search: 382/103
• US Class Current: 382/103
• CPC Class Codes: B61L 17/00 Switching systems for classification yards
• CPC Class Codes: B61L 23/00 Control, warning or like safety means along the route or between vehicles or trains
• CPC Class Codes: B61L 23/041 Obstacle detection
• CPC Class Codes: B61L 25/021 Measuring and recording of train speed
• CPC Class Codes: B61L 25/025 Absolute localisation, e.g. providing geodetic coordinates
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 7/0008 checking presence/absence
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats

Abstract

A method, data processing system, apparatus, and computer program product for monitoring objects. A plurality of images of an area is received. An object in the area is identified from the plurality of images. A plurality of points in a region within the area is identified from a first image in the plurality of images. The plurality of points has a fixed relationship with each other and the region. The object in the area is monitored to determine whether the object has entered the region. A determination that the object has not entered the region is made in response to identifying an absence of a number of the plurality of points in a second image in the plurality of images.

Claims

1. A method for monitoring objects, the method comprising:

receiving a plurality of images of an area;

identifying an object in the area from the plurality of images;

identifying a plurality of points in a region within the area from a first image in the plurality of images, wherein the plurality of points has a fixed relationship with each other and the region;

monitoring the object in the area to determine whether the object has entered the region; and

responsive to identifying an absence of a number of the plurality of points in a second image in the plurality of images, determining, using a processor unit, that a train has entered the region rather than the object.

2. The method of claim 1, wherein responsive to identifying the absence of the number of the plurality of points in the second image in the plurality of images, determining that the object has not entered the region comprises:

determining whether the number of the plurality of points that are absent from the second image exceeds a threshold number; and

responsive to determining that the number of the plurality of points that are absent from the second image exceeds the threshold number, determining that the object has not entered the region.

3. The method of claim 2 further comprising:

identifying a type of the object from the plurality of images;

identifying a size of the object based on the type of the object; and

selecting the threshold number based on a distance between points in the plurality of points and the size of the object.

4. The method of claim 1, wherein the region includes a portion of a set of train tracks, and wherein responsive to identifying the absence of the number of the plurality of points in the second image in the plurality of images, determining that the object has not entered the region comprises:

determining whether a first pattern formed by the number of the plurality of points that are absent from the second image matches a second pattern formed by a train entering the area on the set of train tracks; and

responsive to determining that the first pattern matches the second pattern, determining that the train has entered the region rather than the object.

5. The method of claim 1, wherein identifying the object in the area from the plurality of images comprises:

comparing images in the plurality of images to identify similarities between the images;

identifying the similarities as a set of background objects;

removing the set of background objects from a received image in the plurality of images;

identifying objects remaining in the received image as a set of foreground objects; and

identifying the object from the set of foreground objects;

the method further comprising;

monitoring the set of foreground objects; and

responsive to identifying that the absence of the number of the plurality of points appear in more than one image in the plurality of images, identifying the train as one of the set of foreground objects rather than one of the set of background objects.

6. The method of claim 1, wherein identifying the plurality of points in the region within the area from the first image in the plurality of images comprises:

identifying a value for texture at a plurality of segments within the first image;

identifying an amount of contrast in texture between a first segment and a second segment adjacent to the first segment in the first image; and

responsive to the amount of contrast in texture between the first segment and the second segment exceeding a threshold amount, identifying a point associated with one of the first segment and the second segment as one of the plurality of points in the region.

7. The method of claim 6, wherein the area is a train station, wherein the region includes a portion of a set of train tracks, and wherein the plurality of points include a plurality of points along a rail on a track in the set of train tracks pictured in the first image.

8. The method of claim 7, wherein the plurality of points form a curve along an edge of the rail pictured in the first image and wherein the plurality of points are identified by detecting the edge of the rail in the first image.

9. The method of claim 1, wherein monitoring the object in the area to determine whether the object has entered the region comprises:

identifying a boundary surrounding the region in the area;

setting a width of the boundary; and

responsive to identifying, from an image in the plurality of images, that the object has crossed the width of the boundary, determining that the object has entered the region.

10. An apparatus for monitoring objects, the apparatus comprising:

an object monitoring system operably connected to a camera, the object monitoring system configured to identify an object in an area from a plurality of images of the area generated by the camera and monitor the object in the area to determine whether the object has entered a region within the area, wherein the region includes a portion of a set of train tracks; and

a train detection system comprising a processing unit and operably connected to the object monitoring system, the train detection system configured to;

identify a plurality of points in a region within the area from a first image in the plurality of images, and

determine that a train has entered the region rather than the object response to identifying an absence of a number of the plurality of points in a second image in the plurality of images, wherein the plurality of points has a fixed relationship with each other and the region,

wherein the object monitoring system is further configured to identify a type of the object from the plurality of images and identify a size of the object based on the type of the object and wherein the train detection system is configured to determine that the train has entered the region by determining whether the number of the plurality of points that are absent from the second image exceeds a threshold number, wherein the threshold number is selected based on a distance between points in the plurality of points and the size of the object.

11. The apparatus of claim 10, wherein the train detection system is configured to determine that the train has entered the region by determining whether a first pattern formed by the number of the plurality of points that are absent from the second image matches a second pattern firmed by the train entering the area on the set of train tracks.

12. The apparatus of claim 10, wherein the train detection system is configured to identify the plurality of points in the region within the area from the first image in the plurality of images by identifying a value for texture at a plurality of segments within the first image, identifying an amount of contrast in texture between a first segment and a second segment adjacent to the first segment in the first image, and identify a point associated with one of the first segment and the second segment as one of the plurality of points in the region in response to the amount of contrast in texture between the first segment and the second segment exceeding a threshold amount.

13. A computer program product for monitoring objects, the computer program product comprising:

a set of computer readable storage devices;

program code, stored on one of the set of computer readable storage devices, configured to receive a plurality of images of an area;

program code, stored on one of the set of computer readable storage devices, configured to identify an object in the area from the plurality of images;

program code, stored on one of the set of computer readable storage devices, configured to identify a plurality of points in a region within the area from a first image in the plurality of images, wherein the plurality of points has a fixed relationship with each other and the region;

program code, stored on one of the set of computer readable storage devices, configured to monitor the object in the area to determine whether the object has entered the region; and

program code, stored on one of the set of computer readable storage devices, configured to determine that a train has entered the region rather than the object in response to identifying an absence of a number of the plurality of points in a second image in the plurality of images.

14. The computer program product of claim 13, wherein the program code configured to determine that the object has not entered the region in response to identifying the absence of the number of the plurality of points in the second image in the plurality of images comprises:

program code, stored on one of the set of computer readable storage devices, configured to determine whether the number of the plurality of points that are absent from the second image exceeds a threshold number; and

program code, stored on one of the set of computer readable storage devices, configured to determine that the object has not entered the region in response to determining that the number of the plurality of points that are absent from the second image exceeds the threshold number.

15. The computer program product of claim 13, wherein the region includes a portion of a set of train tracks, and wherein the program code configured to determine that the object has not entered the region in response to identifying the absence of the number of the plurality of points in the second image in the plurality of images comprises:

program code, stored on one of the set of computer readable storage devices, configured to determine whether a first pattern formed by the number of the plurality of points that are absent from the second image matches a second pattern formed by a train entering the area on the set of train tracks; and

program code, stored on one of the set of computer readable storage devices, configured to determine that the train has entered the region rather than the object in response to determining that the first pattern matches the second pattern.

16. The computer program product of claim 13, wherein the program code configured to identify the plurality of points in the region within the area from the first image in the plurality of images comprises:

program code, stored on one of the set of computer readable storage devices, configured to identify a value for texture at a plurality of segments within the first image;

program code, stored on one of the set of computer readable storage devices, configured to identify an amount of contrast in texture between a first segment and a second segment adjacent to the first segment in the first image; and

program code, stored on one of the set of computer readable storage devices, configured to identify a point associated with one of the first segment and the second segment as one of the plurality of points in the region in response to the amount of contrast in texture between the first segment and the second segment exceeding a threshold amount.

17. A data processing system for monitoring objects;

the data processing system comprising;

a bus system;

a storage device connected to the bus system, wherein the storage device includes program code; and

a processor unit connected to the bus system, wherein the processor unit is configured to execute the program code to receive a plurality of images of an area, identify an object in the area from the plurality of images, identify a plurality of points in a region within the area from a first image in the plurality of images, wherein the plurality of points has a fixed relationship with each other and the region, monitor the object in the area to determine whether the object has entered the region, and determine that a train has entered the region rather than the object in response to identifying an absence of a number of the plurality of points in a second image in the plurality of images.

18. The data processing system of claim 17, wherein in executing the program code to determine that the object has not entered the region in response to identifying the absence of the number of the plurality of points in the second image in the plurality of images the processor unit is further configured to execute the program code to determine whether the number of the plurality of points that are absent from the second image exceeds a threshold number and determine that the object has not entered the region in response to determining that the number of the plurality of points that are absent from the second image exceeds the threshold number.

19. The data processing system of claim 17, wherein the region includes a portion of a set of train tracks, and wherein in executing the program code to determine that the object has not entered the region in response to identifying the absence of the number of the plurality of points in the second image in the plurality of images the processor unit is further configured to determine whether a first pattern formed by the number of the plurality of points that are absent from the second image matches a second pattern formed by a train entering the area on the set of train tracks and determine that the train has entered the region rather than the object in response to determining that the first pattern matches the second pattern.

20. The data processing system of claim 17, wherein in executing the program code to identify the plurality of points in the region within the area from the first image in the plurality of images the processor unit is further configured to identify a value for texture at a plurality of segments within the first image, identify an amount of contrast in texture between a first segment and a second segment adjacent to the first segment in the first image, and identify a point associated with one of the first segment and the second segment as one of the plurality of points in the region in response to the amount of contrast in texture between the first segment and the second segment exceeding a threshold amount.