초록 ▼

In trains a dragging air hose can snag an object on the tracks and break. A broken air hose can lead to an emergency stop or an accident caused by faulty brakes. The present invention mounts a video camera opposite a lit screen, thereby contrasting the air hose profile against a white screen in all

In trains a dragging air hose can snag an object on the tracks and break. A broken air hose can lead to an emergency stop or an accident caused by faulty brakes. The present invention mounts a video camera opposite a lit screen, thereby contrasting the air hose profile against a white screen in all weather and lighting conditions. Machine vision algorithms locate the air hose in a captured image from a moving train, and compute its lowest point. An alarm condition is sent to a remote location. An optional car gap detector is laser based. It helps align the car gap with the data capture for the camera. Another subsystem can include a car coupler height detection algorithm.

대표청구항 ▼

We claim: 1. A method to detect a low or broken air hose between passing cars of a railroad train, the method comprising: providing a light uniform backgound parallel to a railroad track at a chosen distance away from the railroad track; mounting a camera on an opposite side of the railroad track;

We claim: 1. A method to detect a low or broken air hose between passing cars of a railroad train, the method comprising: providing a light uniform backgound parallel to a railroad track at a chosen distance away from the railroad track; mounting a camera on an opposite side of the railroad track; aiming the camera at the light uniform background; capturing an image of a gap between two railroad cars; and using a machine vision algorithm to detect a low or broken air hose. 2. The method of claim 1 including the step of using a laser based gap detector to coordinate the capturing of an image. 3. The method of claim 1 including the step of transmitting an alarm signal when a low or broken air hose is detected. 4. The method of claim 1 including the step of using a searchlight aimed at a front of a screen for lighting the screen as the means for providing a light uniform background. 5. The method of claim 1 including the step of using a searchlight aimed at a back of the screen for lighting a screen as the means for providing a light uniform background. 6. The method of claim 1 including the step of using a light box for a screen as the means for providing a light uniform background. 7. The method of claim 1 including the step of using a control computer to capture the image of a gap, and to execute the machine vision algorithm using a vertical line shape edge finder tool. 8. The method of claim 7 including the step of using a train presence wheel sensor and a wheel counter to determine which car has a low or broken air hose. 9. The method of claim 1 including the step of using an algorithm to detect a height of a car coupler. 10. The method of claim 1 including the step of using pattern recognition techniques on a sequence of voltage pulses corresponding to a wheel position of the railroad cars in the train to detect the gaps. 11. The method of claim 1 using specific image patterns generated by the light uniform background that are recognized by the vision algorithm to identify the gap. 12. The method of claim 7 including the step of using a railroad car radio identification tag reader and a wheel counter to determine which gap has a low or broken air hose. 13. The method of claim 1, wherein the light uniform background further comprises a background object of uniform light appearance, including a side of a building or a wall. 14. The method of claim 7, wherein the vertical line shape edge finder tool includes a step of moving a vertical line shape edge finder tool horizontally and vertically across the gap image to detect an air hose in the image by a predetermined width dimension of the air hose. 15. The method of claim 9, wherein the algorithm includes a step of moving a vertical line shape edge finder tool horizontally and vertically across the gap image to detect coupler shanks in the image by their predetermined height dimension. 16. A method to detect a low or broken air hose between passing cars of a railroad train, the method comprising: selecting a planar structure parallel to a railroad track at a chosen distance away from the railroad track; providing light to the planar structure; mounting a camera on an opposite side of the railroad track; aiming the camera at the planar structure; capturing an image of a gap between two railroad cars; and using a machine vision algorithm to detect a low or broken air hose. 17. A system to detect a low or broken air hose between passing cars of a railroad train, the system comprising: means for providing a light uniform background parallel to a railroad track functioning to provide a lighted background to contrast a gap between two railroad cars; means for mounting a camera on an opposite side of the railroad track and aiming the camera at the light uniform background functioning to capture an image of the gap between two railroad cars; and means for using a machine vision algorithm functioning to detect a low or broken air hose via an analysis of the image of the gap. 18. The system of claim 17, wherein the machine vision algorithm further comprises a vertical line shape edge finder tool. 19. The system of claim 17, wherein the machine vision algorithm further comprises a height algorithm to detect a height of a car coupler. 20. The system of claim 17 further comprising a railroad car radio identification tag reader means functioning to determine which gap has a low or broken air hose. 21. The system of claim 18, wherein the vertical line shape edge finder tool includes a step of moving a vertical line shape edge finder tool horizontally and vertically across the gap image to detect an air hose in the image by a predetermined width dimension of the air hose. 22. The method of claim 19, wherein the height algorithm includes a step of moving a vertical line shape edge finder tool horizontally and vertically across the gap image to detect coupler shanks in the image by their predetermined height dimension.