摘要
针对无人机遥感图像中的亮度、尺度、图像重叠等问题,利用BBF优化算法设计无人机遥感图像配准方法;根据无人机遥感成像原理,获取无人机遥感图像作为配准对象,通过图像融合、校正等步骤,完成初始遥感图像的预处理;提取无人机遥感图像的SIFT特征点和轮廓特征,利用BBF优化算法搜索当前图像与参考图像之间的特征匹配对,完成图像的特征匹配;最终通过粗匹配、精匹配和消除错误匹配点3个步骤,得出无人机遥感图像的配准结果;通过效果测试实验得出结论:优化设计方法的特征匹配点准确率为5.15%,漏配率为3.75%,显著优于传统配准算法。
Aiming at the problems of brightness,scale and image overlap in UAV remote sensing images,the best bin first(BBF)optimization algorithm is used to design the UAV remote sensing image registration method.According to UAV remote sensing imaging principle,the UAV remote sensing image is obtained as a registration object,and the initial remote sensing image is preprocessed by the image fusion and correction.The SIFT feature points and contour features of UAV remote sensing images are extracted,the BBF optimization algorithm is adopted to search the feature matching pairs between current image and reference image to complete the feature matching of the image.Finally,through three steps of rough matching,fine matching and eliminating wrong matching points,the registration results of UAV remote sensing images are obtained.The results show that the accuracy of feature matching points of the optimization method is 5.15%and the missing matching rate is 3.75%,which is significantly better than that of traditional registration algorithms.
作者
金红华
JIN Honghua(Yanbian Branch,Jilin Radio and Television University,Yanji 133001,China)
出处
《计算机测量与控制》
2024年第7期232-237,245,共7页
Computer Measurement &Control
基金
中国管理科学研究院教育科学研究所研究课题(KJCX16276)。
关键词
BBF优化算法
无人机遥感图像
图像配准
BF optimization algorithm
UAV remote sensing image
image registration
