Elter, in partnership with Hikvision, has created an innovative solution for detecting graffiti on trains and other vehicles, delivering major time and cost savings for operators. Hikvision interviewed Thomas Anglo, CEO and co-founder of Elter, to find out what's so special about the new graffiti-busting solution, and how the company's partnership with Hikvision is driving AI innovation for the transportation and logistics industries.
Thomas Anglo: First, it's important to mention that some people love graffiti, some hate it, and some don't have strong feelings about it either way. But for train operators, it can be a major challenge for a number of reasons. First, many public transport providers are publicly funded, so keeping trains, buses, rolling stock, and other assets clean is basically respecting public property. Second, private operators need to protect their assets, and to ensure that their brand is displayed cleanly and clearly on trains and other assets – with no graffiti getting in the way. Third, the cost of cleaning graffiti off walls and vehicles is currently huge. In the UK, for example, the total cost of cleaning up graffiti is estimated at £100 million a year1. These costs can be decreased by a faster intervention as the paint removal needs less working time and chemical product if processed within the first 48 hours.
Thomas Anglo: It's always difficult to prevent graffiti because organizations don't have the human resources to watch and protect their assets 24x7. With trains and other vehicles, you also have the added complexity of distributed fleets, with trains parked at night in sidings over a large geographical area. This makes it impossible to monitor and protect trains against graffiti 24x7. As an added challenge, trains are always on the move, which makes it difficult for staff to carry out graffiti checks and to keep track of any affected trains. Even if an operator knows some trains have graffiti on them, it's hard to know which ones, and how extensive the damage is.
Thomas Anglo: We've developed an intelligent AI algorithm that compares passing trains and carriages to a template of a clean and properly painted train. In this way, the algorithm can detect any trains that have been affected by graffiti. What's more, the algorithm can accurately calculate the surface area that has been affected by the graffiti, which allows the train operator to estimate cleaning costs. This AI algorithm is the intelligent 'heart' of our graffiti-busting solution.
Thomas Anglo: Our solution is called TPP Graffiti Spotter. It uses two video cameras with built-in motion detectors on either side of the train track, which allows us to initiate recording when a train comes into the camera's 'field of vision'. Every train engine and carriage has a Radio Frequency Identification (RFID) chip on board, with a unique ID number associated with it, which allows us to identify it as it passes the cameras. The video footage of affected trains is processed by our AI algorithm especially trained to detect and track any kind of graffiti. Additionally, the algorithm detects the extent of the area affected by graffiti and sends this information to the operator and to the contractor responsible for cleaning the graffiti. The operator can access all of this information via a web app which shows the percentage of the fleet affected by graffiti, the total surface area affected, and the specific details of graffiti on each train and carriage.
Thomas Anglo: We chose Hikvision from multiple vendors for three key reasons:
1. Hikvision cameras give us highly reliable motion detection that is unaffected by rain and other adverse conditions. This allows us to capture video footage of trains and carriages reliably, and to record the details of any carriages affected by graffiti.
2. Hikvision cameras offer super high-resolution imaging with great wide-angle capabilities, which allows us to capture an entire carriage within the camera's 'field of vision'.
3. Hikvision provides very good quality imaging in its cameras with very low levels of 'deformation', which means that images are not distorted when trains are in motion. This makes them ideal for the graffiti-detection use case.
Thomas Anglo: About the time we decided to use Hikvision cameras for the TPP Graffiti Spotter solution, Hikvision reached out to us to address a request from one of its own customers also looking for a solution for identifying graffiti – but in another context. It was at this point that we were invited to become a Hikvision partner.
Thomas Anglo: It's great for Elter to be partnering with Hikvision because of the company's industry leading position and global presence. As a small company, this partnership adds a huge amount of credibility and market presence to our innovative AI solutions, and it allows us to start conversations with Hikvision's huge installed base of transportation and logistics customers as well.
Elter's partnership with Hikvision has been a great experience. We conduct all our business in a spirit of kindness and generosity, and it feels like a genuine win-win, even down to this blog and other joint marketing activities we're working on together.
Thomas Anglo: We're currently collaborating with Hikvision experts and sales teams to develop a new range of vehicles inspection services that automatically detect the condition of truck trailers as they enter and leave logistics parks, or containers as they are loaded or unloaded at ports. This solution will allow logistics park owners, shippers, port authorities and others to determine responsibility, and liability, if goods are damaged in transit.
The solutions we're currently developing in collaboration with Hikvision will support our vision for helping transportation and logistics companies automate inspections and monitoring, achieve major cost and time savings, and – critically – to do more with less.
Subscribe to our email newsletter to get the latest, trending content from Hikvision