RazorCam: Embedded Platform for Image Processing

Main computing FPGA as host for a Trenz Electronic GigaBee XC6SLX which is a FPGA micromodule integrating a leading-edge Xilinx Spartan-6 LX FPGA, Gigabit Ethernet transceiver (physical layer), two independent banks of 16-bit-wide 128 MB DDR3 SDRAM.This board has a connection with the motherboard module.TFT display based on a 320 x 240 thin film transistor LCD with serial peripheral interface.SD card slot as mean of external storage Camera Head Modules as host for image sensor.One is an infrared image sensor module and a second one is a digital image sensor module.The infrared image sensor module is based on Tau 640 video camera from FLIR commercial Systems.The digital image sensor module is based on MT9V034, wide VGA CMOS digital image sensor from Aptina. RazorCam ModulesThe RazorCam project is the development of a cooperative smart camera system to develop a flexible and extensible hardware/software environment to prototype and verify video applications.The fisrt step of this project was to develop a hardware platform primarily targeted at real-time image processing.The developed system is capable of streaming image data from a camera headboard, through a Spartan6-XC6SLX45 for processing and analysis.It offers a host of real world interfaces including uart and Ethernet connectivity.Linux is used as the embedded operating system on the microblaze (hard core inside FPGA) as it offers a solid, familiar platform for development with a feature-rich toolchain.In video streaming applications where large amount of data must be processed in real time, this traditional system architecture is no longer viable because the bus becomes a bottleneck.The programmability and the seamless use of hardware accelerators in image processing application is insured through the design and implementation of a streaming data interface.By porting the Intel's OpenCV computer vision library to the system, software developers do not need to recompile their code in order to use the platform.