Amazon Rekognition Workplace Safety

Automate Personal Protective Equipment (PPE) detection to improve workplace safety practices

Safety hazards can exist in every workplace in many different forms: sharp edges, falling objects, flying sparks, chemicals, noise, and a myriad of other potentially dangerous situations. Safety regulators such as Occupational Safety and Health Administration (OSHA) and European Commission often require that businesses protect their employees and customers from hazards that can cause injury by providing them personal protective equipment (PPE) and ensuring their use. Across several industries such as manufacturing, food processing, chemical, healthcare, energy, and logistics improving workplace safety is usually a top priority. In addition, due to the COVID-19 pandemic, wearing PPE in public places has become important to reduce the spread of virus. However, even when people do their best to follow PPE guidelines, sometimes they can inadvertently forget to wear PPE or not realize it is required in the area they are present in. This puts their safety at potential risk and opens the business to possible regulatory compliance issues. Today, businesses rely on site supervisors or superintendents to individually check and remind all the people present in the designated areas to wear PPE, which is not reliable, effective, or cost-efficient at scale.

With Amazon Rekognition PPE detection, customers can analyze images from their on-premises cameras across all locations to automatically detect if persons in the images are wearing the required PPE such as face covers, hand covers, and head covers. Using these PPE detection results, customers can trigger timely alarms or notifications to remind people to wear PPE before or during their presence in a hazardous area to help improve or maintain everyone’s safety. They can also aggregate the PPE detection results and analyze them by time and place to identify how safety warnings or training practices can be improved or generate reports for use during regulatory audits. Note that this feature does not perform facial analysis or facial comparison and cannot identify the detected persons or match their identities.

Benefits

Automate PPE detection at scale

Augment manual checks with automated PPE detection. Analyze images from cameras across all your on-premises sites to detect if employees and customers are wearing PPE where required.

Reduce human and financial risk

Alert or notify employees and customers about missing PPE in time to prevent lapses and improve everyone’s safety. Maintain PPE detection records to comply with occupational safety regulations and reduce the risk of penalties or fines.

Improve safety practices

Store and analyze PPE detection results by different sites and plants to prioritize additional hazard warning signage or safety trainings. Generate detailed PPE detection reports by using AWS Glue, Amazon Athena, and Amazon Quicksight.

Features

Quick summary per image

Simply supply a list of required protective equipment (such as face cover or face cover and head cover) and a minimum confidence threshold (such as 80%) to receive a consolidated summary list of Persons with Required PPE, Persons without Required PPE, and Persons Indeterminate from an image. This reduces the amount of code you need to write for overall counts or find person references in the image to further drill down.

PPE worn predictions

Just detecting the presence of PPE in an image is not very useful. It is important to detect if the PPE is worn by the customer or employee. With Amazon Rekognition PPE detection, you get predictions on whether the protective equipment is covering the corresponding body part. For example, if nose is covered by face cover, head is covered by head cover, and hands are covered by hand covers. This can help you filter out detections of PPE that are not on the person at the right corresponding body part.

Per-person details

Receive full fidelity analysis response including person detection confidence and bounding box (up to 15 persons per image), body part detection confidences, protective equipment detection confidences and bounding boxes, and coversbodypart detection boolean values and confidences. This provides you with the granularity and flexibility to apply business-specific image annotation or notification rules based on each body part, protective equipment, or coversbodypart confidence scores.

Custom PPE detection

If you need to detect a PPE other than face cover, head cover, and hand cover, you can use Amazon Rekognition Custom Labels to detect high-visibility vests, safety goggles or another PPE from your own environment. Simply upload labeled images to train your custom ML model and start detecting. No ML expertise is required. Visit custom PPE detection github repo to learn more

Customers

Omlet (Mobisocial)
“Providing a safe environment for students, teachers, and staff is one of our top priorities. With the COVID-19 challenge, we needed a solution to enable safer classroom and campus experience by making it easy for our staff to check if students and teachers are wearing face masks. With Amazon Rekognition PPE detection, we were able to build a campus wide Virtual Health and Safety Officer that accurately identifies when faculty and students wear face masks for campus, institute building, and classroom entry, as well as remind them in a friendly way to put their mask back on in case they have removed it. Amazon Rekognition PPE made it very easy for us to get started with a pre-trained PPE detection model, saving us valuable time and money that we would have otherwise spent collecting, labeling, and training our own models to work across a variety of environment."

Cyrus Wong, Senior Lecturer, Department of Information Technology (IT) of the Hong Kong Institute of Vocational Education (Lee Wai Lee)

Omlet (Mobisocial)
"Amazon Rekognition PPE detection solved one of our big challenges in this difficult time where we strive to deliver food safely to our customers. By using this technology in our delivery mobile application, we can now automatically check that our food delivery employees are wearing a face mask as they pick up orders from the kitchen and deliver them. Currently in beta, the new version of our delivery application with PPE detection furthers our commitment to food safety standards."

Amit Gupta, CTO, Rebel Foods

Omlet (Mobisocial)
"Amazon Rekognition PPE detection feature provides much higher accuracy compared to other systems we’ve tested. Our customers within industries such as retail, construction, and logistics that require workplace safety solutions are requesting PPE detection and Amazon Rekognition PPE detection allowed us to plug this feature into our application code and get it working straight out of the box.”

James West, CEO, Videoloft

Omlet (Mobisocial)
"Our clients in retail, industrial, and smart cities are thrilled about the new automated PPE detection capability from Amazon Rekognition as it helps them abide by health and safety guidelines, and provide quick response to safety challenges. VXG is leveraging Amazon Rekognition to provide a holistic, customized solution tailored to these organization’s specific needs, with the ability to scale and connect tens of thousands of cameras, and create events/alerts for PPE compliance."

Yaro Lisitsyn, Co-Founder & CEO, VXG

Partners

SmugMug
“Automating the anonymous detection of Personal Protective Equipment (PPE) is important to our customers so they can improve their safety processes and better comply with safety regulations. In addition, the COVID-19 pandemic has made automated face mask detection more important than ever. We are excited that the new Amazon Rekognition PPE detection feature is now available to our customers as part of Deloitte’s computer vision platform “Horizon by Deloitte”.

Mike LeBlanc, Partner, Deloitte

Omlet (Mobisocial)
"PPE detection with Amazon Rekognition provides a valuable tool in promoting worker safety that can be readily integrated into existing video streaming workloads to add additional insights and business value. This enables us to better and more quickly enable our customers to identify PPE violations and provide feedback on ways to improve safety operations."

Charles Burden, Head of Business Development, TensorIoT

CBS
"By leveraging the new Amazon Rekognition PPE detection, we have been able to quickly add and launch additional safety and policy compliance features within “Workwatch”, our back to work solution for businesses that are aiming to safely bring back customers and employees to their business premises.”

Carl Krupitzer, CEO, ThingLogix

Learn more about Amazon Rekognition pricing

Visit the pricing page
Ready to build?
Get started with Amazon Rekognition
Have more questions?
Contact us