PYNQ FPGA Hardware implementation of LeNet-5-Based Traffic Sign Recognition Application

Computer Vision is one of the most flagship fields in the last decade. Meanwhile, embedded boards (FPGAs) represent the technological advancement trends by including software processor and programmable logic units at the same FPGA core. However, the fusion of the embedded system and computer vision provides the nascent embedded vision field. This domain occupied all areas, even future Industry 4.0, by exploiting them in control and management tasks. Otherwise, embedded vision with artificial intelligence plays an important role in many vital applications such as advanced driver-assistance systems Otherwise, embedded vision with artificial intelligence plays an important role in many vital applications such as advanced driver-assistance systems in smart transportation context. This paper falls within this range, by implementing LeNet-5 Model-based Traffic Sign Recognition application on Xilinx embedded FPGA system. Implementation results prove that our Co-design prototype, including Processing System (PS) and Programmable Logic (PL) parts, achieves best performance in terms of hardware cost and run-time execution.