High‑Performance FPGA Accelerator for Embedded Vision with MIPI Cameras
Vision Components has engineered a cutting‑edge FPGA accelerator that performs real‑time edge pre‑processing of image data for embedded vision projects powered by MIPI camera modules. The compact board supports multiple MIPI‑CSI‑2 inputs and outputs, enabling complex image processing and data analysis directly on the edge.
Leveraging a powerful, fully programmable FPGA, the accelerator can merge feeds from several MIPI cameras, execute sophisticated algorithms, and handle demanding computing operations. It will ship early in 2022 as an open‑hardware platform, allowing customers to program the FPGA and deploy demo applications immediately.
In a subsequent release, Vision Components will provide proprietary FPGA designs tailored to specific tasks such as color conversion, 1D barcode recognition, and epipolar correction. Plans also include AI acceleration support. Developers can integrate the module into their electronic designs with the same simplicity as a standard MIPI camera, transferring pre‑processed image results to a host CPU via MIPI for further processing.

Vision Components expands its MIPI camera portfolio with new modules. A 50 mm × 50 mm board incorporates the Melexis MLX75027 time‑of‑flight sensor, featuring 10 µm × 10 µm pixels and DepthSense technology for high‑resolution, high‑contrast 3‑D imaging. The ToF MIPI module and four additional boards with Sony Pregius S sensors—offering global shutters and up to 12 MP resolution—will be available for order later in 2021.
Future additions will include MIPI camera modules for short‑wave infrared (SWIR) imaging, employing diverse sensor technologies to deliver cost‑effective SWIR solutions.
Embedded
- IC Technology & Microcontrollers: The Cornerstone of Modern Embedded Systems
- Embedded Vision Drives Innovation Across Industries
- Ambarella Launches CV5 AI Vision Processor: 8K at 30fps in <2W, Redefining Power‑Efficient Imaging
- ST Microelectronics Showcases 28 nm FD‑SOI MCUs with Embedded PCM for Powertrain, ADAS, and Electrification
- Cervoz Ultra‑Thin M.2 NVMe SSDs: High‑Performance Storage for Industrial Edge Applications
- Optimizing AI Models for Efficient Embedded Deployment
- FPGA Implementation Languages: From VHDL and Verilog to High‑Level C/C++
- Getting Started with FPGA Development: Building a 4‑Bit Adder on Xilinx Artix‑7
- Embedded FPGA (eFPGA) Technology: Revolutionizing ASIC and SoC Design
- 9 Key Factors for Selecting and Designing Board‑Level Machine Vision Cameras