On Semiconductor has licensed a vision embedded computing platform from Ceva for its automotive advanced driver assistance system (ADAS) product lines.
Ceva's DSP technology gives new image processing capabilities and enables On Semiconductor to incorporate embedded intelligence and machine learning into its ADAS roadmap.
‘The automotive industry requires cost- and power-efficient vision-based ADAS solutions to address the growing end-customer demand and safety regulations across all tiers of the automotive industry,’ said Ross Jatou, vice president and general manager of the automotive solutions division of On Semiconductor.
Market research firm Strategy Analytics has predicted demand for automotive cameras to reach almost 200 million units in 2023, while Yole Développement expects advanced imaging technology for automotive applications to be worth $7.3 billion by 2021. Yole believes that ADAS will account for 51 per cent of this revenue.
Automotive imaging applications are computationally intensive. Embedded processors are able to handle low-light image processing and run powerful deep neural networks that can provide the accuracy and performance of future active safety systems. These reasons are what led On Semiconductor to select Ceva's imaging and vision platform to augment its ADAS product offerings.
Ceva's XM DSP processors include a hybrid architecture composed of scalar and vector DSP processors coupled with a comprehensive application development kit to streamline software deployment. They address the intensive processing requirements found in ADAS and reduce the power consumption of the overall systems.
