计算机科学
接头(建筑物)
分辨率(逻辑)
光学
计算机图形学(图像)
人工智能
物理
建筑工程
工程类
作者
Xijun Zhao,Bin Fan,Zijie Ma,Shuo Zhong,Jianying Chen,Tao Zhang,Haibing Su
标识
DOI:10.1016/j.optlaseng.2024.108308
摘要
Multi-order diffraction lenses can focus different wavelengths at the same focal point by utilizing various diffraction orders. However, due to material dispersion, the focal position shifts when the incident wavelength moves away from the primary wavelength, resulting in limited spectral bandwidth. This paper presents a computationally optimized multi-order diffractive lens design and a related image reconstruction algorithm, enabling broadband high-resolution imaging with a single element. We introduce a new type of lens, the Spectral Multi-Order Diffractive Lens (SMODLs), which redistributes the energy across different diffractive orders to achieve relatively high focusing performance within specified wavelength bands. Subsequently. The relatively sharp spectral images generated by SMODL within these specified bands serve as priors for reconstructing images across the full spectrum. A prototype lens with a 40 mm aperture and a 320 mm focal length was designed and fabricated. The prior images' wavebands were designated as 525 nm to 575 nm and 675 nm to 725 nm, with peak Strehl ratios of 0.47 and 0.35, respectively. Finally, grayscale images within the 500∼800 nm wavelength range were reconstructed. Experimental results confirm the effectiveness of our approach, achieving high-resolution imaging across a 300 nm bandwidth with a single diffractive element.
科研通智能强力驱动
Strongly Powered by AbleSci AI