XAIS and EyeVi complete first fully automated survey in London

The 30th June 2021 proved to be a landmark day for EyeVi and XAIS as an innovative project with Barnet Council came to a successful conclusion – resulting in 761 km worth of road data being analyzed within 10 datasets and 2,654 areas of interest being detected. The collaboration was of significant value for both parties: EyeVi was able to demonstrate the capabilities of its fully automated defect detection process and XAIS elevated its digital survey capabilities to the next level. Here’s the story of how it all began and how the project went…

Need help with automated defects detection? We’ve got the answer!

As an innovative and ambitious company that offers digital road and asset management services, XAIS was looking for a partner that could offer cutting-edge and high-quality road defects detection services. At the same time, EyeVi was looking for a partner that was operating in the field of road infrastructure management – such as a road consultant or a company offering road asset management – with a need for digital data processing. At the beginning of 2021, the executives of the two companies met and it turned out to be a match made in heaven.  

The first chance to collaborate was on a project for Barnet Council in London. The aim was to digitize the road network in Barnet, detect road defects using Artificial Intelligence (AI), and then calculate the costs of roads maintenance in the area. All objectives were achieved thanks to an intelligent and efficient approach that involved EyeVi running the entire process automatically in the UK with zero manual corrections to the data – a significant accomplishment.

So, what exactly have we done with Barnet?

We delivered a significant result for Barnet – digitizing all of the roads, automatically annotating road defects that belong to eight different categories, and calculating road maintenance costs. Anyone familiar with the field understands the scale and importance of this achievement – which involved expertly planned and perfectly executed processes.

First, to undertake analysis, you need data. So, to collect data of the highest quality, EyeVi sent a range of equipment to XAIS for initial testing and calibration. This included a 360° panoramic camera, LIDAR scanner, GNSS/INS system, power inverter, and data collection computer with software.  

Data from the initial test runs was then checked by EyeVi to make sure that everything was working perfectly. After getting the green light from EyeVi, XAIS started the live data capture.  

When the data capture process was finished, XAIS uploaded the raw data to EyeVi’s servers and the fully automated data processing began. The data was initially run through EyeVi DataFlow – an automated data processing pipeline – where georeferenced panoramic images and HD orthophotos were produced. Next, the processed data was input into the AI module which was tasked with detecting all potential defects.  After the AI finished its work, the results were converted into a data table that the XAIS team used for their calculations. XAIS has its own algorithm that merges the data, analyzes the network, and pulls out specific information as required by its clients. In addition, XAIS adds other relevant information, such as the possible existence of other defects and patching levels.

Providing added value

Without a doubt, the Barnet project was a huge success for XAIS and EyeVi with a chance to prove that their innovative collaboration could deliver a range of positive outcomes.

The fully automated process provided by EyeVi helped XAIS and its client save considerable time and money when compared with the traditional method of sending engineers out in the field to undertake a manual visual survey of the network.  

Alongside these savings, the images and data captured during the initial collection phase can be used to benchmark future surveys and reused for other purposes – such as detecting the number and condition of traffic lights or other highways assets.

Finally, the chance to collect, process, and analyze significant volumes of data presented opportunities to enhance ways of working still further – pushing technology to achieve exceptional levels of accuracy and efficiency.

From the data collected in London, XAIS now has a pipeline of 6 Highway Authorities, which are using the data as above and for use in scheme development using XA(c) Scheme Assembler tool.

A brilliant outcome and an exciting future ahead.