复杂时空环境的异分辨率红外/可见光图像非刚性配准与异表征融合研究
结题报告
批准号:
62003247
项目类别:
青年科学基金项目
资助金额:
24.0 万元
负责人:
樊凡
依托单位:
学科分类:
自动化检测技术与装置
结题年份:
2023
批准年份:
2020
项目状态:
已结题
项目参与者:
樊凡
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
本项目针对无人机对地感知在应用拓展过程中遇到的瓶颈问题展开研究,即在复杂时空环境下,如何通过图像融合手段,提升图像的探测和识别能力。具体要解决复杂时空环境下的异分辨率的红外/可见光图像非刚性配准与异信息表征融合问题,旨在形成目标信息显著、高分辨率、纹理细节丰富的无人机融合成像。具体研究:复杂时空环境下的红外与可见光高效高精准配准,使复杂时空环境下的红外与可见光图像对应像素表征的物理位置相同;研究异分辨率条件下红外强度与可见光梯度融合方法,将高分辨率可见光图像与低分辨率红外图像在同一数学表达中进行建模求解,使高分辨率可见光图像中表征的梯度纹理信息充分准确地传递给红外图像;搭建实验验证平台,将研究的理论与方法进行算法集成,开展仿真实验和真实实验验证融合成像性能,优化算法理论。本项目将形成一套完整的应用于无人机对地观测的红外可见光融合成像理论方法,可有力促进红外成像在灾害救援等领域的应用。
英文摘要
This project focuses on the bottleneck encountered in the application expansion of unmanned aerial vehicle (UAV) remote sensing to the ground, that is, how to improve the image detection and recognition ability by image fusion in the complex spatiotemporal environment. Specifically, the problem of non-rigid registration and heterogeneous characteristic fusion of different resolution infrared and visible images in complex spatiotemporal environment should be solved. It aims to form the fusion image of UAV with significant targets, high resolution and rich texture details. Specifically, this project researches: 1) high accurate non-rigid registration of the infrared and visible images in complex spatiotemporal environment, so that their corresponding pixels are at the same physical position; 2) the fusion method of the infrared intensity and the visible gradient under the condition of different resolution, where the high-resolution visible image and the low-resolution infrared image are modeled and solved in the same mathematical expression, so that the gradient texture information represented in the high-resolution visible image can be fully and accurately transferred to the infrared image; 3) the experimental verification platform building, the theory and method integration into the hardware platform, validation of the fusion performance by simulation and experiment, and optimization of the algorithm. This project will form a complete set of infrared and visible fusion methods applied to UAV sensing, which can effectively promote the application of infrared imaging in disaster relief and other fields.
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:10.1109/jas.2022.105686
发表时间:2022-07-01
期刊:IEEE-CAA JOURNAL OF AUTOMATICA SINICA
影响因子:11.8
作者:Ma, Jiayi;Tang, Linfeng;Ma, Yong
通讯作者:Ma, Yong
DOI:10.1109/tii.2021.3118682
发表时间:2021
期刊:IEEE Transactions on Industrial Informatics
影响因子:12.3
作者:Jiayi Ma;Wenjing Gao;Yong Ma;Jun Huang;Fan Fan
通讯作者:Fan Fan
DOI:10.1109/lgrs.2022.3196433
发表时间:2022
期刊:IEEE Geoscience and Remote Sensing Letters
影响因子:4.8
作者:Zhao-bing Qiu;Yong Ma;Fan Fan;Jun Huang;Lang Wu
通讯作者:Lang Wu
DOI:10.1016/j.inffus.2022.07.013
发表时间:2022-08-20
期刊:INFORMATION FUSION
影响因子:18.6
作者:Le, Zhuliang;Huang, Jun;Ma, Jiayi
通讯作者:Ma, Jiayi
Hyperspectral Anomaly Detection With Robust Graph Autoencoders
使用鲁棒图自动编码器进行高光谱异常检测
DOI:10.1109/tgrs.2021.3097097
发表时间:2021-07-21
期刊:IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
影响因子:8.2
作者:Fan, Ganghui;Ma, Yong;Ma, Jiayi
通讯作者:Ma, Jiayi
国内基金
海外基金