Computer Vision for Railway Asset Management

Railway Signal Detection

By placing a dash camera in front of the engines, we will be able to identify the condition and the working status of the railway signals. A report will be generated if the wrong signal is displayed or the signals are off. 

· Working status of tunnel lights:

By processing the dash camera video, we will also be able to gather information about the lighting conditions inside the tunnel, if any are not working and needs replacement or if more lights are required to be installed.

· Signboard Detection:

We will be able to monitor and track the condition of the signboards present along the railway tracks. A report will be generated if any signboard is missing, broken or bent. 

We will also be able to track the condition of any linear assets such as fencing, barriers, etc. If any part is damaged, missing or bent, a report will be generated.

· Accident Prone Areas:

The data from the dash camera can be used to find out accident-prone regions along the tracks. There could be a region where a lot of animals cross the track. Also, in the village areas, there could be people crossing the tracks often. This data could be used to undertake preventive measures in those regions.

· Road Barricades:

We will be able to find out unmanned road barricades. If any of the road barricades are open when the train crosses, a report will be generated with a picture of the open barricades.

· Violation of Rules:

We will be able to identify if anyone has breached the reserved areas around the railway tracks. For example, there could be temporary stalls set up near the tracks, nomad tents, or even vegetation growing near the tracks. 

We could keep track of the speed of trains. By pre-storing the max speed allowed in any area, we will be able to find out if a train overspeeds, and a report will be generated.

· Minimum Lighting Conditions:

We will be able to identify regions where lighting conditions are minimum, for example, a crossing near a village. We can use this data for preventive action and could install caution lights for speeding trains in those regions. 

· Railway Track Condition:

We will be able to monitor the condition of the railway track. For example, any foreign object on the track will be identified and reported. Also, a report will be generated if the cement connectors between the tracks are damaged, bent or not aligned.

Station Surveillance: 

· Water Leakages:

We will be able to monitor water leakages in the pipes along the railway tracks. We will also to able to identify water leakages from other sources in the Railway Station.

· Stray Animals:

From Cameras placed in the railway stations, we will be able to report any stray animal present there. This data can be used for preventive action.

· Violation of Rules (Station):

We will be able to find out if any person violates the station rules. For example, if people are not using the overbridges for crossing the tracks, if people try to enter restricted areas, if people try to enter the station from unauthorized entry points, etc.

· Crowd Analysis:

We will be able to report the number of people present at the railway station at any given point in time. This data can be used for detailed analysis.

· Threat Detection:

We will be able to alert the officials if any unattended bags are found in the railway station.

· Hawkers Detection:

We can generate a report of hawkers in the station. We will be able to know the time of their visits and movements. 

· Stalls Activity Detection:

We will be able to identify the opening & closing times of the stalls. If the stalls are unmanned or opened late, a report will be generated with respective images. 

· Trash Detection:

We will be able to detect trash thrown in the railway station. The pattern of trash disposal can also be identified, and this data can be used to place trash cans in suitable areas.