Amazon Rekognition now detects Personal Protective Equipment (PPE) such as face covers, head covers, and hand covers on persons in images

Posted on: Oct 16, 2020

Amazon Rekognition is a deep-learning-based image and video analysis service that can identify objects, text, scenes, as well as support content moderation by detecting inappropriate content. Starting today, Amazon Rekognition can analyze images to detect if persons in the images are wearing Personal Protective Equipment (PPE) such as face covers (face masks), hand covers (gloves), and head covers (helmets or hard hats).

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 the 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 businesses 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 multiple locations to automatically detect if persons in the images are wearing the required PPE. When customers analyze an image using Amazon Rekognition PPE detection, for each person detected in the image, they receive confidence scores with bounding boxes for each item of protective equipment detected (face cover, hand covers, and head cover) and Boolean responses (true or false) with confidence scores for whether each detected item of protective equipment covers the corresponding body part (nose, hands, and head). They can also 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 per-image summary of Persons with Required PPE, Persons without Required PPE, and Persons Indeterminate. Using these PPE detection results, customers can trigger timely alarms or notifications that remind people to wear PPE before or during their presence in a hazardous area in order 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 and generate reports for use during regulatory audits. Note that Amazon Rekognition PPE detection does not perform facial recognition or facial comparison and cannot identify the detected persons.

Amazon Rekognition PPE detection is now generally available and can be used in all regions where Amazon Rekognition is supported. To learn more or get started, please visit the product webpage, read our blog, refer to our documentation, and download the latest AWS SDK. To try the feature with your images, you can sign into the Amazon Rekognition console and use the Amazon Rekognition PPE detection demo.