The arduous process of developing optical lenses for imaging systems is set to undergo a dramatic change thanks to a groundbreaking AI-powered method developed by researchers at King Abdullah University of Science and Technology (KAUST).
This innovative approach, called DeepLens, automates the design process and promises to significantly reduce development time and costs while improving image quality for mobile phone cameras.
DeepLens: A new era in lens design
(Photo: wtrsnvc_ from Unsplash)
A group of researchers at a Saudi Arabian university has developed a groundbreaking AI model to help develop optical lenses for mobile phones.
The DeepLens design method, developed by researchers Xinge Yang, Qiang Fu and Wolfgang Heidrich at KAUST, uses a groundbreaking technique known as “curriculum learning,” according to Interesting Engineering.
This structured, iterative approach carefully considers key imaging system parameters such as resolution, aperture and field of view to achieve optimal lens design.
Implementing curriculum learning in AI
Just as humans learn complex tasks step by step, curriculum learning enables artificial intelligence (AI) systems to tackle complicated problems step by step.
Before a person can walk or dance, for example, they must first learn to crawl and walk. The DeepLens method works in a similar way: the AI breaks down the complex task of designing a sophisticated lens system into manageable milestones, gradually increasing the requirements for resolution, aperture size and field of view.
Unlike traditional methods that require a human-designed template as a starting point, the DeepLens approach autonomously generates its own design for a composite optical system. This system consists of multiple refractive lens elements, each with its own tailored shapes and properties to ensure optimal performance, according to SciTech Daily.
Related article: AI vision is not that superior to human vision, new study shows
Advantages over conventional methods
According to Xinge Yang, one of the developers of the DeepLens method, conventional automated methods can only slightly optimize current optical designs.
The researcher said this could be of great benefit as it could reduce “months of manual work by experienced engineers” to just a single day.
The effectiveness of the DeepLens approach has already been demonstrated in the creation of both classical optical designs and computational lenses with extended depth of field.
In one example, the method was used to design a mobile phone-sized lens system with a large field of view, featuring lens elements with highly aspherical surfaces and a short back focal length.
The design and optical performance of this classic six-element imaging system were analyzed as the system evolved to meet design specifications.
Expanding the application area of DeepLens
Currently, the DeepLens method is limited to refractive lens elements. However, the research team at KAUST is actively working to extend the approach to hybrid optical systems that combine refractive lenses with diffractive optics and metal lenses. This advancement could further miniaturize imaging systems and unlock new possibilities such as spectral cameras and joint color depth imaging.
“This will further miniaturize imaging systems and open up new possibilities such as spectral cameras and shared color depth imaging,” concluded Yang.
The innovative approach leverages the power of AI and curriculum learning and promises to streamline the design process, reduce costs and pave the way for the next generation of imaging systems.
Whether in mobile phone cameras or other advanced imaging applications, the impact of this technology will be far-reaching and set new standards for performance and capability in the industry.
You can read the full study in Nature Communications by clicking here.
Also read: Tech giants clash with India over OTT regulation
ⓒ 2024 TECHTIMES.com All rights reserved. No reproduction without permission.
