How Image Recognition Technology Can Save the Construction Industry $1.6 Trillion Annually

Category: AI Insights

Published date: 04.12.2020

Read time: 6 min

Construction projects can be viewed as large jigsaw puzzles that have to be put together piece by piece at exactly the right time. The timing of each phase is important in order to make sure that the entire project will be done on time. However, as projects get larger and more complex it becomes increasingly difficult to manage each stage which can cause the completion date to be delayed. In fact, these sorts of mistakes and delays are very expensive and cost the construction industry $1.6 trillion every year. AI is helping to reduce or even eliminate some of these costs with the help of image recognition technology. 

The Practical Applications of Image Recognition Software

Let’s say that your construction company is working on a project to create a building with 1,500 rooms. Usually, you would only have 5 managers overseeing the construction which is not rough to keep track of all the details and nuances involved in the project. This is where AI-powered image recognition software can help keep the project on schedule. A new British-Israeli startup, Buildots, has created such technology and it is already being used by two of the largest construction companies in Europe, including the British giant Wates. 

When managers tour a site once or twice a week, there will be a GoPro camera mounted on top of their helmets which will take video footage of the project and upload it into image recognition software. This technology will compare the status of thousands of objects on site down to the smallest details such as whether all of the objects have been bolted properly, all of the electrical sockets are in the right place and many others by creating a digital replica of the building. The AI technology tracks the location of the camera inside the building to within a few centimeters, allowing it to identify the coordinates of each object in the frame. According to the Founder and CEO of Buildots, Roy Danon, this product can track the status of about 150,000 different objects, many times every week. 

Image recognition requires a lot of quality data annotation work to be performed in order for the system to accurately identify each object. In this case, the training dataset would need to be larger since the computer needs to identify not only the object but also its status. This means that human data annotators would have to annotate thousands and thousands of images and label each item and the state it’s in according to a gradation scale. This can be something as simple as not started, in progress, complete, or any other status required by the construction company. 

AI is Reducing the Workload for Site Inspectors 

Site inspectors need to make sure that the building plans and specifications are being followed correctly and manage staff and subcontractors on building sites. They don’t have time for repetitive tasks such as checking to see if the walls have been painted or if there is an unnecessary hole drilled somewhere. This is something that can easily be automated and AI is taking care of these mundane tasks so that site inspectors and other managers can focus on their core functions. Also, since all of the changes are logged automatically, there is no need to walk around with a spreadsheet or schedule to track everything manually.

The Importance of AI in the Time of COVID-19

We have written previously about how the pace of automation is speeding up in response to the COVID-19 pandemic and the subsequent lockdown restrictions to ensure social distancing and eliminating the need for people to leave their home or office without necessity. The image recognition software we described above is also helping keep employees safe since it allows managers to keep track of projects remotely. When only the essential workers were allowed to work on the site, their helmets were equipped with cameras that took video footage of the project and uploaded it into the system. 

Mindy Support Offers Comprehensive Data Annotation Services 

While Buildots requires a specific type of data annotation, there are many other ones used to create cutting-edge AI and machine learning projects. Mindy Support can help you with all of your data annotation needs by assembling a team of data annotators for you. Regardless of the size or complexity of your project, we will source and recruit the necessary amount of candidates quickly and provide them with the necessary training so they can hit the ground running. We are one of the largest BPO providers in Eastern Europe with a successful track record of actualizing all kinds of data annotation projects. 

Posted by Il’ya Dudkin


    Stay connected with our latest updates by subscribing to our newsletter.

      ✔︎ Well done! You're on the list now