ecchelmet Computer Vision is giving EHS the edge by
counting the uncountable
3.30 am. A small fire breaks out at the L&T Innovation Campus project site in Manappakam, Chennai. In normal circumstances, it would have taken a long while before EHS In-charge T Jeyakasi would have known about it. However, thanks to ‘Computer Vision’ (CV), the fire was instantly detected, personnel alerted, and the danger averted in double quick time. That is just one example of how digitalization is helping the project team to ‘count the uncountable’, easily detect Unsafe Acts and Unsafe Conditions, and take the necessary precautionary measures in time to prevent serious incidents.
Demystifying CV
CV is an aspect of AI that gives the system cognitive capabilities. By feeding hundreds of different images and video feeds (directly from CCTV) into the system, it is trained to recognize and classify images that have elements of unsafe acts or unsafe conditions, with a variety of machine learning techniques and algorithms to achieve this. Once the system is trained, it is configured to recognize and alert the appropriate staff about them, and the system keeps learning from every new image that it processes.
“Image analytics platforms these days can analyse video footage from video surveillance cameras that are already set up at most modern workplaces,” remarks Mahesh Aravind Chikodi, Head – Digital Solutions & Analytics. “They are simple to deploy and require little to no additional hardware investment, and the advances in machine learning are making CV applications more accurate and dependable. In fact, the application of CV can increase productivity, improve safety, reduce costs, and lead to more well-informed planning decisions,” he adds, extolling the benefits of this digital solution. Image analysis is, therefore, certainly making inroads into construction sites, and none more so than at the Innovation Campus project site. CV is a powerful and automated tool that can extract images and video information from sites to effectively monitor and identify risky factors that are the result of unsafe behaviour of construction workmen.
Image analytics platforms these days can analyse video footage from video surveillance cameras that are already set up at most modern workplaces. The application of CV can increase productivity, improve safety, reduce costs, and lead to more well-informed planning decisions.
Mahesh Aravind Chikodi
Head – Digital Solutions & Analytics
List of EHS Use Cases
- Safety – PPE (hardhat, jacket, shoes)
- Harness detection/hooking
- Detecting wrong movement (transport movement on site as per logistic plan)
- Detecting speed violations
- Pedestrian path and movement
- Movement of equipment near power lines
- P&M equipment being moved in restricted areas
- Detecting the gathering of crowds
- Detecting people moving in restricted areas or movement of workmen in non-working areas at site
- Detecting if a person collapses
- Detecting fire hazards (handling of gas cylinders or combustive materials)
- Edge barricading
- Detecting persons standing, leaning, or sitting on the edge at height
- Material kept at the edge
- Detecting loose or protruding material
- Detecting abandoned objects
- Detecting vehicles, number plates, and optical character recognition (OCR)
- Assault detection
- Unauthorized entry at sensitive sites – nuclear plants, refineries, etc.
- Fire, smoke detection – air quality monitoring for confined spaces, mines, welding
- Posture detection
“Unsafe behaviour of construction workmen is one of the main causes for safety accidents at construction sites,”
points out Project Manager Singaram Vadivazhagan. “To reduce the incidence of such accidents and improve safety performance, there is a need to identify the risky factors by monitoring behaviour. At our Innovation Towers site, we have found that CV is the answer to this problem that is helping us to monitor site performance and get alerted instantly of any lapses.”
Unsafe behaviour of construction workmen is one of the main causes for safety accidents at construction sites. At our Innovation Towers site, we have found that CV is the answer to this problem that is helping us to monitor site performance and get alerted instantly of any lapses.
Singaram Vadivazhagan
Project Manage
Adding more intelligence to CV
“While the platform is inherently robust and intelligent, it was critical to customize it for our specific requirements,” informs Mahesh, “for which we drew up an exhaustive list of EHS use cases.”
I can check and correct people working at height without a harness or those who have dropped a tool. I am warned of wrong vehicle and people movement, people wandering into restricted areas, the functioning and condition of machines, hazardous situations, intrusions, and so much more.
T Jeyakasi
EHS In-charge
CV is heaven-sent for Jeyakasi, for the 8 cameras that electronically watch his site alert him on a variety of infringements. “I can check and correct people working at height without a harness or those who have dropped a tool,” he explains. “I am warned of wrong vehicle and people movement, people wandering into restricted areas, the functioning and condition of machines, hazardous situations, intrusions, and so much more,” he smiles gratefully.
A boon not just for EHS; CV makes eminent business sense
Monitoring PPE compliance: Most accidents and incidents at construction sites can be attributed to people not wearing proper PPE. CV can address that by instantly warning the security officer if anyone is entering the site without hardhats, high-visibility jackets, gloves, masks, footwear, or any other protective equipment worn by employees and workmen at site. The captured video footage can be stored for later incident analysis. “CV does not take coffee breaks, never sleeps; it works constantly, monitoring the site,” smiles Mahesh.
Tracking material quantities: CV can track the quantities of building materials stored at site and can trigger advance alerts if quantities are running low.
Optimizing workflow: At large construction sites, inefficient movement of workers can result in significant time cost.
CV creates spaghetti diagrams to identify workmen movement trajectories; it can check for longer travel paths, movement bottlenecks, and optimize on-site material storage.
Measuring manual cycle duration: CV can identify the duration of manual cycles, most relevant for our labourintensive industry. The software can measure how long a particular task takes (for instance, laying a line of pavement tiles). Like the BIM 3D progress tracking model, CV can produce real-life project completion estimates to check whether the construction progress is in line with the project timeline.
Preventing unauthorized access: CV solutions are already being utilized in many worksites for access control. By identifying license plates, it can prevent unauthorized vehicle access. Taking things further, the technology utilizes face recognition to automatically identify the entry of unauthorized persons.
While the fact that CV technology is enormously beneficial is undeniable, the Innovation Campus site is reaping its benefits, with several other sites from across ICs on the threshold of adoption. A few projects ready to convert are the Navi Mumbai Airport, Development of Ayodhya, IKEA GMP Package 2, JRS India, Prestige City, Prestige DB Turf, Redevelopment of PMCH, SCB Medical Expansion, LTR Phase 3, C65 Commercial Tower, High Speed
Rail – C4 Package, RRTS – Track Works & Yard, and Blast Furnace #5 for JVML, Ballari.
The sooner projects take the plunge, the better it will be to dramatically improve our collective EHS performance. What are we waiting for?