Scope: To develop a next-gen industry grade machine vision camera with improved functionalities that can read codes on the widest range of materials
Solution:
- Designed and developed modular FPGA-based vision processing unit (VPU) to enhance the imaging engine with edge analytics
- Pre-functional prototype of the VPU for feasibility study, analysis, and architecture
- Hardware: Adaptor board design, interface Camera, CPU and VPU boards
- Software: Yocto Linux BSP porting, driver development, integration, and board bring up
- Integrating, implementing and configuring drivers (PMIC, Ethernet, RPMsg, pin muxing, camera sensor, V4L2, Wi-Fi, Bluetooth, PTP, etc.)
- Camera driver support enhancements and support for dual-camera setup
- Adding driver support for new peripherals (OLED display support, temperature sensor, accelerometer)
Impact:
- Real-time advanced image processing and analytics on the edge
- Modular engine enabling rapid development of multiple product lines
- Reduced time-to-market by 8 months
- Provided higher percentage of factory automation and automated real-time checks for defect identification and other quality KPIs to customer’s customer in Automotive, Logistics, Consumer Electronics, and Pharmaceutical industries