Transportation Risk Management
Bringing AI and Computer Vision to Transportation Risk Management. How Intel and IncluIT are using the Intel Distribution of OpenVINO toolkit.
The two companies understood that Internet of Things (IoT) solutions represent an unprecedented opportunity to transform the way trucking businesses deal with today’s transportation issues. The benefits were many and included improved operational efficiency, reduced costs, and decreased risk.
To enhance efficiency on the road, IncluIT turned to a roster of new and emerging tools spanning edge computing, artificial intelligence (AI), computer vision, and an open-source Intel solution designed to help bring it all together, the Intel® Distribution of OpenVINO™ toolkit.
Harnessing the Intel® Distribution of OpenVINO™ toolkit
- Allows driver monitoring to prevent and reduce risks by identifying and inferencing potentially dangerous situations.
- Helps speed development of applications and solutions that emulate human vision.
- Enables deep learning inference at the edge while also scaling across Intel Vision Products to provide: The Intel® Deep Learning Deployment toolkit, pretrained models and samples and optimized functions for OpenCV and OpenVX.
- Supports heterogeneous execution across computer vision accelerators -from CPUs and GPUs to the Intel® Neural Compute Stick and FPGAs- using a common API.
The POC: Fleet management and driver management
IncluIT’s Driver Management System is built on an architecture based on edge processing and cloud streaming, the system tracks driver behavior, alerting the driver of potentially dangerous situations. The POC uses the model detection capabilities of the Intel Distribution of the OpenVINO™ toolkit, enabling users to read faces with great accuracy. To do that the toolkit relies on two detection models: facial recognition and gesture detection.
- The Intel Distribution of OpenVINO™ toolkit
- Intel® Core™ CPU with integrated graphics
- Intel Neural Compute Stick and Intel® Myriad™ X VPU
- AWS Dynamo Database
- AWS ELK
- Standard USB Video Class (UVC) camera