Camera manufacturer and image processing expert Allied Vision announced it has become an official member of the NVIDIA Jetson ecosystem and is giving developers on these powerful AI embedded computing systems access to Allied Vision camera modules for industrial image processing. With the release of the Vimba MIPI CSI-2 driver, Allied Vision’s open-source driver supports NVIDIA’s Jetson TX2 System-on-Module (SoM) with Linux for Tegra, and is now available on GitHub.
The Alvium camera series combines the advantages of embedded sensor modules with the performance of industrial cameras for image processing: extensive functions for image correction and optimization, a large selection of state-of-the-art sensors, intelligent energy management and a cost-optimized and compact design. It opens new possibilities for the design of AI-enabled embedded systems. The Vimba driver for NVIDIA Jetson supports all current and future Alvium camera modules with MIPI CSI-2 interface, no matter which sensor the camera module uses. If an application system based on the Jetson platform is developed, the same driver can be used, even if another Alvium camera with a sensor of a different resolution is selected later. This makes it easy to develop and prototype new systems and replace the camera. Comprehensive documentation and support further facilitate system integration and simplify prototyping.
Relief of embedded processor
Equipped with an integrated Image Signal Processor (ISP) and extensive image processing library (IPL), the Alvium camera can assume image processing tasks and be configured natively via the Video4Linux interface, allowing the internal ISP of the Jetson TX2 board to be relieved. Further image processing tasks, which usually require GPU resources, can also be performed by the camera. The computing capacity and speed are thus free for demanding applications such as scientific analysis, the development of new internet services, autonomous vehicles, or other AI applications.
“With the Vimba MIPI CSI-2 Driver for NVIDIA Jetson and its adapter boards, system designers now have a combination of a powerful SoM, easy-to-use software, and a camera model that can be used to implement technologically advanced applications,” stated Mark Schenk, product manager software at Allied Vision.
“We welcome Allied Vision as a new member of the NVIDIA ecosystem. The Alvium camera series offers users of the Jetson platform unprecedented possibilities to bring AI to the edge,” said Murali Gopalakrishna, head of product management, Autonomous Machines at NVIDIA.
The Jetson TX2 driver is just the first step in the ecosystem collaboration. Allied Vision will soon release additional drivers for other products in the NVIDIA Jetson family, including Jetson AGX Xavier and Jetson Nano. Allied Vision’s partner Antmicro has already demonstrated Allied Vision Alvium camera modules working in combination with the Jetson product portfolio.
“We are proud to be part of the NVIDIA Jetson ecosystem and look forward to expanding our Jetson platform offerings to provide developers of Jetson-based systems with new opportunities in machine vision,” concluded Sebastian Günther, embedded strategy manager at Allied Vision.