AI Case Study

Skanska identifies potential jobsite safety hazards in real-time using machine learning to automatically tag videos and photos with infractions occurring

Skanska has been using Smartvid.io, an industrial photo and video analytics platform to identify potential safety hazards in real time. The technology uses machine learning-powered technology to analyse the visual content of the footage and automatically tag these hazars using "SmartTags".

manage media captured from diverse sources, such as its drone program or field progress walks.
Once in Smartvid.io, photos and videos are assigned labels (called “SmartTags”) based on their visual content, such as a crane, as well as any accompanying audio content like voiceovers or recordings from field personnel, by leveraging machine learning-powered technology.
Next, the SmartTagging engine automatically tagged potential safety hazards like PPE violations (SmartTags included ‘Missing Hard Hat’ and ‘Missing Safety Vest’).

Industry

Industrials

Construction And Engineering

Project Overview

"Smartvid.io provides a single location for all photo and video content from current and past projects. Once in Smartvid.io, photos and videos are assigned labels (called “SmartTags”) based on their visual content, such as a crane, as well as any accompanying audio content like voiceovers or recordings from field personnel, by leveraging machine learning-powered technology. When Smartvid.io introduced SmartTags for detecting common safety risks like missing personal protective equipment (PPE), Skanska’s innovation and safety teams began exploring
how it could apply the technology to its safety programs. Skanska tested Smartvid.io’s SmartTag functionality to help with an emerging best practice: capturing narrated videos of project walks at key project phases, highlighting proper implementation of safety practices and potential risks. Using GoPro cameras and/
or the Smartvid.io mobile app via an iPhone, site safety managers took videos narrating observed examples of safety best practices
and potential hazards.

Skanska then began testing Smartvid.io’s integration with Autodesk’s BIM 360, which teams have used since 2008 for document management, quality control, and other eld activities. Once integrated, jobsite photos stored in the BIM 360 database - whether added by project managers, engineers, or subcontractors - automatically synced to Smartvid.io by project. Next, the SmartTagging engine automatically tagged potential safety hazards like PPE violations (SmartTags included ‘Missing Hard Hat’ and ‘Missing Safety Vest’)."

Reported Results

Specifics undisclosed, however the vendor states that the automatic aggregating and indexing of existing photo reduced manual labour time and will be used to improve jobsite safety training.

Technology

"The Smartvid.io Smart Tagging engine, nicknamed “VINNIE”, leverages machine learning (also called “Deep learning”) technology powered by NVIDIA. NVIDIA’s Quadro M6000 and Quadro GP100 boards power the deep learning training and model production hardware used by Smartvid.io to build their vision and speech models tailored to the needs of construction. When Skanska uploads photos and videos to VINNIE, NVIDIA’s GPUs power the Amazon cloud services doing the real time analysis."

Function

Human Resources

Health And Safety

Background

Skanska is a construction and development company which "began piloting Smartvid.io, an industrial photo and video analytics platform in 2016 to manage media captured from diverse sources,
such as its drone program or field progress walks".

Benefits

Data

Photographs and videos taken from cameras and mobile apps.