Ghost imaging, inspired by the concept of quantum entanglement, is a new opto-electronic imaging technique, which involves different scientific fields. Single-pixel detection, which is the major distinguishing feature of this imaging technique, allows ghost imaging of multi-wavelength detection and non-local imaging. Ghost imaging goes beyond the limitations of conventional imaging techniques, expands the applications of imaging and has great potential. However, single-pixel detection also causes problems, the target information is spatially and temporally intermingled during the detection process and the time to reconstruct the image is long. These problems, which limit the dimension and speed of the information acquisition in ghost imaging, are the most urgent problems wanted to be solved by researchers working in related fields..Based on our previous research and ongoing collaborations, this project will explore the mechanism behind the extraction of three dimensional (3D) object information from the spatially and temporally intermingled signals, develop an efficient sampling method using orthonormal bases and a real-time 3D reconstruction algorithm based on adaptive compressive sensing, propose an accurate imaging system by utilizing digital microscanning to ghost imaging. The project will realize 3D measurement with millimeter resolution and sub-second update rate of the proposed photon counting ghost imaging system and verify them through experiment. The project aims to achieve a major breakthrough in the fundamental theory and technical application of real-time 3D ghost imaging, to offer more and faster information for ghost imaging, and to provide a solid foundation for future ghost imaging research in China.
鬼成像技术起源于量子纠缠概念,是一种涉及多学科交叉的新型光电成像技术。单像素探测是其最显著特征,使鬼成像技术具有复合谱探测和非定域成像的原理性优势,突破了传统成像的技术限制,扩展了光电成像的应用范围,潜力巨大。然而,单像素探测体制存在目标信息时空混叠、成像实时性差等问题,严重限制了鬼成像获取信息的维度和速度,是相关领域科学工作者亟待攻克的难点。.本项目在已有研究成果和国内外合作的基础上,探索时空混叠的探测信号中目标三维信息提取的机理,研究基于标准正交基的高效采样和自适应压缩感知的实时成像技术,研究基于数字微扫描与鬼成像相结合的高精度图像重建技术,实现光子计数鬼成像三维测量的毫米级成像分辨率和亚秒级成像更新率;争取在三维实时鬼成像系统的基础理论和技术发展方面取得突破,使鬼成像获取更多、更快的信息,为鬼成像技术的进一步发展提供坚实基础。
本项目针对鬼成像技术中单像素探测体制存在目标信息时空混叠、成像实时性差等问题,项目研究团队在已有研究成果和国内外合作的基础上,揭示了时空混叠的探测信号中目标三维信息提取的机理,提出了基于数字微扫描与鬼成像相结合的高精度图像重建技术,实现了基于标准正交基的高效采样和自适应压缩感知的实时成像技术,搭建了基于光子计数的三维鬼成像实验系统,验证了系统的毫米级成像分辨率和亚秒级成像更新率。具体研究内容与相应成果为:.(1)揭示了时空混叠的光子探测信号中目标三维信息提取的机理,确定了三维鬼成像系统的总体方案,建立了基于光子计数探测的综合噪声模型,研究成果发表《Optica》等SCI论文5篇。.(2)提出了基于非均匀采样下成像的Hadamard矩阵二维表征方法;设计了非均匀调制矩阵,提高了关键视场图像的分辨率。研究成果发表《Science Advances》等SCI论文3篇,其中ESI高被引论文1篇。.(3)提出了基于DMD调制的高速非均匀微扫描方法,并分析了微扫描探测对于三维精度的影响,进一步提高了三维目标场景的探测精度。研究成果发表《Photonics Research》等SCI论文4篇。.(4)完成了基于高速的LED阵列结构光照明技术,实现了2.5MHz的结构光照明速度,较现有基于DMD的结构光照明技术(22kHz)提高了两个数量级。研究成果发表《Optics Express》等SCI论文3篇,其中ESI高被引论文1篇。.本项目经过四年的研究,按项目任务书规定的预期成果,搭建了基于光子计数探测和非均匀采样的三维鬼成像实验系统,实验结果表明,该系统达到0.5mm的纵向距离分辨率和122Hz的成像更新率。本项目相关研究成果发表SCI论文15篇,培养博士后1人,博士7人、硕士8人。
{{i.achievement_title}}
数据更新时间:2023-05-31
基于 Kronecker 压缩感知的宽带 MIMO 雷达高分辨三维成像
混采地震数据高效高精度分离处理方法研究进展
自组装短肽SciobioⅡ对关节软骨损伤修复过程的探究
不同分子分型乳腺癌的多模态超声特征和临床病理对照研究
下调SNHG16对胃癌细胞HGC-27细胞周期的影响
基于光子轨道角动量纠缠的压缩“鬼”成像研究
基于光子统计特性的深度学习鬼成像理论与算法研究
基于压缩感知理论的光子计数成像方法研究
基于空间关联技术的高速多光束光子计数激光成像