A compact and low-loss multimode waveguide bend plays a significant role in multimode channels and highdensity on-chip optical interconnection architectures,and it has become a key component of photonic integrated chi...A compact and low-loss multimode waveguide bend plays a significant role in multimode channels and highdensity on-chip optical interconnection architectures,and it has become a key component of photonic integrated chips.Here,we propose an inverse design method based on staged optimization of high-order Bezier curves with high degrees of freedom,which effectively overcomes the optimization limitations of traditional geometric curve design.Using this approach,we demonstrate an ultra-compact 90°multimode waveguide bend on a 220 nm silicon-on-insulator(SOl)platform,featuring an effective bending radius as small as 9μm and supporting four TE modes.Furthermore,the bend is extended to arbitrary angle interconnects,with 60°,120°,and 180°configurations as examples,significantly enhancing the flexibility and adaptability of on-chip compact multimode interconnections.Simulation results show that 90°bend exhibits excellent performance at 1550 nm with excess losses below 0.038 dB and crosstalk below-30 dB.The proposed design was further fabricated and experimentally characterized.The maximum measured excess loss is 0.13 dB,and the inter-mode crosstalk is all below-25 dB at 1550 nm.This device combines ultra-compact footprint,low loss,and excellent scalability,suitable for high-density on-chip interconnects.展开更多
Imaging through scattering media faces a critical challenge:deep-learning-based methods inherently suppress high-frequency speckle information,limiting the recovery of fine textures and edges.To overcome this spectral...Imaging through scattering media faces a critical challenge:deep-learning-based methods inherently suppress high-frequency speckle information,limiting the recovery of fine textures and edges.To overcome this spectral bias,we introduce the concept of the relative speckle frequency domain(RsFD),which redefines high-frequency features as learnable,adaptive components via frequency-domain decomposition.We demonstrate that independently processing generalized high-frequency speckle components enables neural networks to capture latent target details previously obscured in conventional approaches.Leveraging this principle,we design FDUnet,a dualbranch network comprising a low-frequency sub-network(Lnet)for global structure reconstruction and a relative high-frequency sub-network(RHnet)dedicated to enhancing textures and edges.Experiments confirm FDUnet's superiority:it outperforms state-of-the-art methods in both visual fidelity and quantitative metrics by +5.9% to 8.7% in SSIM and+5.4 to 7.9 dB in PSNR across diverse datasets(MNIST,Fashion-MNIST,FERET).These enhancements translate into notable improvements in the preservation of textures and edges,especially exhibiting exceptional robustness to multimode fiber perturbations.This work bridges the gap between physical priors and neural network learning,unlocking new potentials for high-fidelity applications,such as biomedical endoscopic imaging,in dynamic scattering environments.展开更多
With the advent of the era of big data,artificial intelligence has attracted continuous attention from all walks of life,and has been widely used in medical image analysis,molecular and material science,language recog...With the advent of the era of big data,artificial intelligence has attracted continuous attention from all walks of life,and has been widely used in medical image analysis,molecular and material science,language recognition and other fields.As the basis of artificial intelligence,the research results of neural network are remarkable.However,due to the inherent defect that electrical signal is easily interfered and the processing speed is proportional to the energy loss,researchers have turned their attention to light,trying to build neural networks in the field of optics,making full use of the parallel processing ability of light to solve the problems of electronic neural networks.After continuous research and development,optical neural network has become the forefront of the world.Here,we mainly introduce the development of this field,summarize and compare some classical researches and algorithm theories,and look forward to the future of optical neural network.展开更多
基金National Natural Science Foundation of China(62305381,62105152,62525509)Fundamental Research Funds for the Central Universities(30919011401,30922010204,30922010718,JSGP202202)+4 种基金Funds of the Key Laboratory of National Defense Science and Technology(6142113210205)Leading Technology of Jiangsu Basic Research Plan(BK20192003)National Defense Pre-Research Foundation of China during the 14th Five-Year Plan Period(D040107)Jiangsu Funding Program for Excellent Postdoctoral TalentPostgraduate Research Practice Innovation Program of Jiangsu Province。
文摘A compact and low-loss multimode waveguide bend plays a significant role in multimode channels and highdensity on-chip optical interconnection architectures,and it has become a key component of photonic integrated chips.Here,we propose an inverse design method based on staged optimization of high-order Bezier curves with high degrees of freedom,which effectively overcomes the optimization limitations of traditional geometric curve design.Using this approach,we demonstrate an ultra-compact 90°multimode waveguide bend on a 220 nm silicon-on-insulator(SOl)platform,featuring an effective bending radius as small as 9μm and supporting four TE modes.Furthermore,the bend is extended to arbitrary angle interconnects,with 60°,120°,and 180°configurations as examples,significantly enhancing the flexibility and adaptability of on-chip compact multimode interconnections.Simulation results show that 90°bend exhibits excellent performance at 1550 nm with excess losses below 0.038 dB and crosstalk below-30 dB.The proposed design was further fabricated and experimentally characterized.The maximum measured excess loss is 0.13 dB,and the inter-mode crosstalk is all below-25 dB at 1550 nm.This device combines ultra-compact footprint,low loss,and excellent scalability,suitable for high-density on-chip interconnects.
基金National Natural Science Foundation of China(62362037)Fundamental Research Funds for the Central Universities(30919011401,30920010001)+3 种基金Natural Science Foundation of Jiangxi Province(20224ACB202011)Jiangsu Province Key Research and Development Project(BE2023817)Hong Kong Research Grant Council(15217721,15125724,C7074-21GF)Hong Kong Polytechnic University(P0045680,P0043485,P0045762,P0049101)。
文摘Imaging through scattering media faces a critical challenge:deep-learning-based methods inherently suppress high-frequency speckle information,limiting the recovery of fine textures and edges.To overcome this spectral bias,we introduce the concept of the relative speckle frequency domain(RsFD),which redefines high-frequency features as learnable,adaptive components via frequency-domain decomposition.We demonstrate that independently processing generalized high-frequency speckle components enables neural networks to capture latent target details previously obscured in conventional approaches.Leveraging this principle,we design FDUnet,a dualbranch network comprising a low-frequency sub-network(Lnet)for global structure reconstruction and a relative high-frequency sub-network(RHnet)dedicated to enhancing textures and edges.Experiments confirm FDUnet's superiority:it outperforms state-of-the-art methods in both visual fidelity and quantitative metrics by +5.9% to 8.7% in SSIM and+5.4 to 7.9 dB in PSNR across diverse datasets(MNIST,Fashion-MNIST,FERET).These enhancements translate into notable improvements in the preservation of textures and edges,especially exhibiting exceptional robustness to multimode fiber perturbations.This work bridges the gap between physical priors and neural network learning,unlocking new potentials for high-fidelity applications,such as biomedical endoscopic imaging,in dynamic scattering environments.
基金supported in part by the National Natural Science Foundation of China under Grant 11773018 and Grant 61727802in part by the Key Research and Development programs in Jiangsu China under Grant BE2018126+1 种基金in part by the Fundamental Research Funds for the Central Universities under Grant 30919011401 and Grant 30920010001in part by the Leading Technology of Jiangsu Basic Research Plan under Grant BK20192003.
文摘With the advent of the era of big data,artificial intelligence has attracted continuous attention from all walks of life,and has been widely used in medical image analysis,molecular and material science,language recognition and other fields.As the basis of artificial intelligence,the research results of neural network are remarkable.However,due to the inherent defect that electrical signal is easily interfered and the processing speed is proportional to the energy loss,researchers have turned their attention to light,trying to build neural networks in the field of optics,making full use of the parallel processing ability of light to solve the problems of electronic neural networks.After continuous research and development,optical neural network has become the forefront of the world.Here,we mainly introduce the development of this field,summarize and compare some classical researches and algorithm theories,and look forward to the future of optical neural network.
