Recent advances in embedded vision have evolved from passive video capturing devices to fully autonomous vision systems. Self-driving cars, drones, or autonomous guided robots require real-time parallel processing, low-latency, and in some cases low current consumption. Multiple camera modules provide surround view and sensor fusion improves the overall vision system, while artificial intelligence and machine learning herald tremendous improvements for recognition and learning tasks in autonomous vision systems.
Building autonomous vision systems grows increasingly challenging, requiring multi-disciplinary expertise in optics, image sensors, computer vision and deep learning. Selecting the right development platform and design methodology is crucial for a successful implementation. Avnet Silica offers embedded vision solutions bundled with tools and reference designs to minimize development time in creating autonomous vision systems.
Requirement of a multi-camera solution
- High data rate handling multiple camera input
- Low latency for real-time execution / fast reaction time
- Low power consumption
- Reliability and safety
- Flexibility and reconfigurability
- Intensive computer vision and machine learning
The high-resolution multi-camera solution consists of 3 hardware core building blocks: camera modules, a multi-sensor FMC module, and a multi-processor SoC or SOM on a carrier board.
|Camera modules||Multi-Camera FMC module||Carrier card|
|ON Semiconductor||Maxim Integrated||Maxim Integrated||Xilinx|
An image pipeline starts at the camera module combining image sensor, lens, control electronics, and an interface. Modules with an implemented image sensor and serializer like MARS, a Modular Automotive Reference System developed by ON Semiconductor, developers can build different combinations of image sensors, coprocessors or ISPs, and communication standards.
The camera module supports high bandwidth and image resolution, and data can be transferred over several meters to the multiprocessor SoC with the GMSL serializer from Maxim Integrated.Read More
Carrier board with Multi-processor SoC or SOM
The carrier board provides the mounting option of the Multi-sensor FMC module and additional modules or cards. It includes also the ZynqUltraScale+ MPSoC device or connector for the Avnet UltraZed-EV SOM.
Several configurations are available to build the multi-camera solutionread more
Multi-sensor FMC module
The multi-sensor FMC module is an interconnection board between the camera modules and the host carrier board. It is not a stand-alone module, but rather a plug-in module designed to interface with FMC-compatible carrier boards.
The Avnet Multi-Camera FMC module supports up to four (4) high definition camera modules using MAXIM Integrated’s GMSL (Gigabit Multimedia Serial Link) technology.Read More