Collie Personal Protective Equipment (PPE) detection will show information of workers wearing PPE in an image. You can use this information to improve workplace safety practices. For example, you can use PPE detection to help determine if workers on a construction site are wearing head covers, or if medical workers are wearing face covers and hand covers. The following image shows some of the types of PPE that can detected.
Types of Personal Protective Equipment
Face cover
Detecting common face covers such as surgical, N95, and masks made of cloth.
Hand cover
Detecting hand covers such as surgical gloves and safety gloves.
Head cover
Detecting hard hats and helmets.
You can optionally request a summary of the PPE items and persons detected in an image. You can specify a list of required protective equipment (face cover, hand cover, or head cover) and a minimum confidence threshold (for example, 80%). The response includes a consolidated per-image identifier (ID) summary of persons with the required PPE, persons without the required PPE, and persons where a determination couldn’t be made.
The summary allows you to quickly answer questions such as How many persons are not wearing face covers? or Is everyone wearing PPE? Each detected person in the summary has a unique ID. You can use the ID find out information such as the bounding box location of a person not wearing PPE.
Drawing bounding boxes around face covers
The example displays a bounding box around detected face covers. The bounding box is green if the face cover fully covers the body part. Otherwise a red bounding box is displayed. As a warning, a yellow bounding box is displayed within the face cover bounding box, if the detection confidence is lower than the specified confidence value. If a face cover is not detected, a red bounding box is drawn around the person.
Collie Personal Protective Equipment (PPE) detection will show information of workers wearing PPE in an image. You can use this information to improve workplace safety practices. For example, you can use PPE detection to help determine if workers on a construction site are wearing head covers, or if medical workers are wearing face covers and hand covers. The following image shows some of the types of PPE that can detected.
Types of Personal Protective Equipment
Face cover
Detecting common face covers such as surgical, N95, and masks made of cloth.
Hand cover
Detecting hand covers such as surgical gloves and safety gloves.
Head cover
Detecting hard hats and helmets.
You can optionally request a summary of the PPE items and persons detected in an image. You can specify a list of required protective equipment (face cover, hand cover, or head cover) and a minimum confidence threshold (for example, 80%). The response includes a consolidated per-image identifier (ID) summary of persons with the required PPE, persons without the required PPE, and persons where a determination couldn’t be made.
The summary allows you to quickly answer questions such as How many persons are not wearing face covers? or Is everyone wearing PPE? Each detected person in the summary has a unique ID. You can use the ID find out information such as the bounding box location of a person not wearing PPE.
Drawing bounding boxes around face covers
The example displays a bounding box around detected face covers. The bounding box is green if the face cover fully covers the body part. Otherwise a red bounding box is displayed. As a warning, a yellow bounding box is displayed within the face cover bounding box, if the detection confidence is lower than the specified confidence value. If a face cover is not detected, a red bounding box is drawn around the person.
The image output is similar to the following.
Reference
Guideline for PPE Detction
See Also
Stranger Detection
People Counting
Video Management System
Recent Posts
Categories
Archives