Access Control for Secure Areas
Reliably recognizing safety helmets and high-visibility vests
Secure identification and classification
The system first identifies people who want to enter the secure area. Then the Deep Learning algorithm classifies objects (in this case the helmets and high-visibility vests), enabling the system to recognize whether the lead person is wearing the required safety gear. The system will only grant access if both the safety helmet and vest are present.
A dashboard displays the evaluated data, showing how many people were allowed or denied access.
Embedded vision system components
The image processing system includes all of the components necessary for the application:
up² board with Movidius Add-on Chipset
Basler dart USB camera module with 5 MP resolution
Lens
Software solution with Deep Learning
The main component of this solution is the application software, which is based on a customized Convolutional Neural Network (CNN). This model uses the latest Deep Learning and on-device processing techniques for a fast and reliable response. The CNN is trained on the host side, while its execution (inference) takes place on the device itself.
This safety system offers:
High reliability and speed
Lower labor costs with improved safety standards
A robust system optimized for image processing
Industry-proven, durable hardware
Various integration options thanks to a cost-efficient, slim design
Extendable neural network for flexibility in deployment
Products for this solution
Looking to implement a comparable solution? These products will help you.
