For decentralized preprocessing of image data in embedded vision projects with MIPI camera modules, Vision Components has developed an FPGA-based accelerator with multiple MIPI CSI-2 inputs and outputs. The board can combine data from multiple MIPI cameras and masters complex algorithms and computational operations thanks to a freely programmable FPGA. The accelerator will initially be available in early 2022 with a completely open FPGA for customer programming and with demo applications. In the next step, a custom FPGA design for specific applications such as color conversion, 1D barcode identification, epipolar correction, etc. is planned. The use of the electronics for AI acceleration is also envisioned.
