项目名称: 像素级多传感器图像融合关键问题研究
项目编号: No.60871096
项目类型: 面上项目
立项/批准年度: 2009
项目学科: 轻工业、手工业
项目作者: 李树涛
作者单位: 湖南大学
项目金额: 35万元
中文摘要: 像素级多传感器图像融合是将同类或异类传感器获取的关于同一场景或目标的图像综合成一幅图像,从而获得对该场景或目标的更为准确、全面、可靠的图像描述,在军事和民用领域都具有广阔的应用前景。本项目以动态数码图像、红外与微光夜视图像为对象,深入研究并形成图像的混合多分辨率分解理论、面向图像融合的图像稀疏表达理论和图像区域分割理论与方法,解决无法配准的动态数码图像融合、符合人眼视觉特性的红外与微光夜视的彩色融合等关键问题,在像素级图像信息融合新理论、新方法的研究中取得突破,研究成果将在军用夜视和数码成像等领域具有重要的应用价值。
中文关键词: 图像融合;混合多分辨率;稀疏表示;多传感器
英文摘要: Multisensor image fusion on pixel-level fuses multiple images on one scene or one object from homogeneous or heterogeneous sensors to obtain more accurate, reliable and comprehensive information. It has wide applications in military and civil fields. This project studies hybrid multiresolution decomposition theory and image sparse representation theory aimed to image fusion, and image segmentation method to solve those cases that images could not be registered and infrared and night vision image fusion. The research objects are dynamic digital images and infrared and night vision images. This project will breakthrough the new theory and methods of image fusion. And research achievements has important values in military night vision and digital cameras.
英文关键词: Image fusion; hybrid multiresolution; sparse representation; multisensor