像素级图像融合方法与融合效果评价  

Pixel Level Image Fusion Methods and Fusion Performance Evaluation

作　　者：夏明革[1] 何友[1] 欧阳文[2] 

机构地区：[1]海军航空工程学院电子工程系,山东烟台264001 [2]海军工程大学兵器工程系,湖北武汉430033

出　　处：《遥感技术与应用》2002年第4期224-229,共6页Remote Sensing Technology and Application

基　　金：高等学校全国优秀博士学位论文作者专项基金 ( 2 0 0 0 36)

摘　　要：图像融合的目的是将同一场景的多幅图像的互补信息合并成一幅新图像 ,以便更好地完成场景进行监视和侦察等任务 ,是在多测度空间综合处理多源图像和图像序列的技术。融合图像更适合人的视觉和便于图像的后续处理 ,如图像分割、特征提取等。介绍了像素级图像融合的几种方法 ,按空间域和变换域对各种方法进行了分类 ,并对各种方法进行了比较 ;融合图像应保留原图像的重要细节信息且不引入虚假信息 ;Image fusion is currently an active research field. The objective of image fusion is to combine information from multiple images of the same scene to achieve inferences that cannot be achieved with a single image or source. Image fusion refers to the techniques that integrate complementary information from multi-image sensor data such that the new images are more suitable for the purpose of human visual perception and the computer-processing tasks such as segmentation, feature extraction, and object recognition. There are two essential requirements for image fusion: (1) pattern conservation: important details of the component images must be preserved in the composite image; and (2) spurious elements avoidance: it must not introduce any new pattern elements or artifacts that could interfere with subsequent image analysis and reconstruction. Pixel level fusion is low-level fusion which uses basic information. In this paper, a few image fusion methods are discussed and compared, and image fusion is classified based on space domain and transform domain. Performance evaluation of multi-sensor image fusion is a key problem in image fusion. This paper discusses performance evaluation of image fusion using cross entropy too.

关 键 词：像素级图像融合方法 融合效果评价 图像代数 金字塔变换 小波变换 交叉熵 计算机视觉 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]