基于图像自适应增强与动态融合算法的数字TDI CMOS高动态范围成像方法研究
结题报告
批准号:
62005280
项目类别:
青年科学基金项目
资助金额:
24.0 万元
负责人:
兰太吉
学科分类:
光学信息获取、显示与处理
结题年份:
2023
批准年份:
2020
项目状态:
已结题
项目参与者:
兰太吉
国基评审专家1V1指导 中标率高出同行96.8%
结合最新热点,提供专业选题建议
深度指导申报书撰写,确保创新可行
指导项目中标800+,快速提高中标率
客服二维码
微信扫码咨询
中文摘要
数字TDI CMOS成像技术凭借其突出的灵活性、经济性,在航天航空光学遥感领域占据日益重要的地位。然而受数字图像信号叠加原理的限制,其较低的成像动态范围是亟待解决的难题。基于前期研究基础,本项目提出从成像机理角度入手,以实时图像处理为手段优化成像链路、改进高动态成像性能。具体通过深入解析CMOS图像传感器信号产生、传递以及数字TDI信号叠加过程,构建任意级数图像特征的评价机制及其随积分级数变化规律的理论模型;以图像最佳评价参数为边界条件建立迭代求解区域自适应增强变换参数的方程;基于图像融合效果最优化确立增强后图像的融合算法及权重参数动态选择策略;开展多类型场景成像实验,对变换参数和理论模型等进行校验与修正,提升算法的普适性与可靠性。本项目的研究工作不仅对提高面阵CMOS的遥感推扫成像应用水平具有重要的理论意义,而且能为推动数字TDI技术在其他高速运动场景中的高性能应用带来广阔的前景。
英文摘要
Digital TDI CMOS imaging technology is playing an increasingly important role in the field of aerospace optical remote sensing because of its outstanding flexibility and economy. However, limited by the signal superposition mechanism, the low dynamic range of digital TDI is a difficult buy vital problem to be solved. Based on the previous research, we propose the idea of optimizing imaging link and improving imaging performance by means of real-time image processing from the perspective of imaging mechanism. Firstly, through deeply analyzing the generation and transmission of CMOS signals and the signal superposition process of digital TDI, the theoretical model of image evaluation and its law of change with TDI-stages is established. Secondly, taking the best evaluation parameter of image as the boundary condition, an iterative algorithm is proposed to solve the equation of the adaptive enhancement transformation parameters to realize the regional adaptive image enhancement. Thirdly, based on the optimization of image fusion effect, the fusion algorithm of enhanced image and the selection strategy of the weighting factor are constructed, which can effectively guide the image fusion process from the theoretical level. Finally, the imaging experiment of multi-type scene is carried out to verify the transformation model and parameters, and to improve the universality and reliability of the algorithm. The research of this project not only has important theoretical significance to improve the application level of push scan imaging of area array CMOS, but also can bring broad prospects for promoting the high-performance application of digital TDI technology in other high-speed moving scenes.
期刊论文列表
专著列表
科研奖励列表
会议论文列表
专利列表
DOI:--
发表时间:2022
期刊:影像科学与光化学
影响因子:--
作者:吴梦飞;薛旭成;兰太吉;徐鑫伟
通讯作者:徐鑫伟
DOI:--
发表时间:2022
期刊:无线电工程
影响因子:--
作者:徐鑫伟;兰太吉;薛旭成;吴梦飞
通讯作者:吴梦飞
DOI:10.3390/rs13020204
发表时间:2021
期刊:Remote. Sens.
影响因子:--
作者:Ting Nie;Liang Huang;Hongxing Liu;Xiansheng Li;Yuchen Zhao;Hangfei Yuan;Xiangyu Song;B. He
通讯作者:Ting Nie;Liang Huang;Hongxing Liu;Xiansheng Li;Yuchen Zhao;Hangfei Yuan;Xiangyu Song;B. He
DOI:10.1364/ao.419457
发表时间:2021
期刊:APPLIED OPTICS
影响因子:1.9
作者:Li Jun-Lin;Yang Yong-Qiang;Lan Tai-Ji;Tang Yan-Fu;Li Hong-Yu;Li Zhong-Ming
通讯作者:Li Zhong-Ming
DOI:10.1007/s12596-022-00941-w
发表时间:2022-08
期刊:Journal of Optics
影响因子:2.1
作者:Junlin Li;Yanfu Tang;Yongqiang Yang;Hongyu Li;Taiji Lan
通讯作者:Junlin Li;Yanfu Tang;Yongqiang Yang;Hongyu Li;Taiji Lan
国内基金
海外基金