摘要
高分辨率遥感图像中飞机目标的检测和识别具有重要的军事和民用价值,针对以往方法易受灰度分布和形态变化及伪装干扰等缺点,提出一种基于视觉词袋模型的高分辨率遥感图像飞机目标检测的新方法。为了精简飞机视觉码本得到最具鉴别力的视觉单词,结合相关性及冗余度分析去除视觉码本中不相关、弱相关以及冗余的视觉单词,选择对飞机目标检测最为重要的视觉单词,减少了计算复杂度,提高了算法的检测性能。
It is difficult to detect the target of aircrafts in high resolution remote sensing ima- ges. So a bag-of visual-words model is proposed. In order to abtain the most discriminative visual words and make visual codebook effective and compact, the visual words that are irrele- vant, weakly irrelevant and redundant must be removed from the visual codebook. The method integrating relevance analysis with redundancy analysis is used to prune out those useless visual words. Finally, the most important visual words are chosen to describe aircrafts in high resolu- tion remote sensing images, which helps reducing the computation in the following test stage and also improving the efficiency.
出处
《数据采集与处理》
CSCD
北大核心
2014年第1期60-65,共6页
Journal of Data Acquisition and Processing
基金
国家自然科学基金(61170200)资助项目
江苏省科技支撑计划(BE2012179)资助项目
关键词
遥感图像
飞机检测
特征选择
视觉词袋
remote sensing imagery aircraft detection feature selection bag-of-visual-words
