摘要
针对传统图像拼接算法存在的缺点,提出一种基于局部熵的全自动图像拼接算法.首先使用Harris角点检测器提取二幅图像的特征点.利用熵精确定位的优点,采用局部最小熵差和匹配点约束关系相结合的双向特征匹配算法匹配特征点.为剔除误匹配点的干扰,利用快速RANSAC算法(PERANSAC)对匹配点求精,估算单应矩阵.最后用三角函数加权平均法进行图像融合,平滑处理,消除接缝,实验验证了本算法性能.
By analyzing the disadvantages of the traditional image mosaic algorithms,an automatic image mosaic algorithm based on local entropy is proposed.At first it uses Harris corner detector to extract the feature points from two images.The algorithm is based on local entropy of precise positioning advantage,the least local entropy and corners＇ constraint relations are used to match interest points.PERANSANC algorithm is used to reject false matches and compute homography matrix.The Triangle Function is used to get smoothing seamless mosaic image.The result shows that the algorithm is robust and effective.
出处
《微电子学与计算机》
CSCD
北大核心
2013年第1期65-68,共4页
Microelectronics & Computer
基金
黑龙江高等教育学会"十二五"教育科学研究规划课题(HGJXH B1110957
HGJXH B2110958)
哈尔滨学院学科发展研究基金项目(HXKQ200602)
关键词
图像配准
角点检测
局部熵差
图像融合
Image registration
corner detector
local entropy difference
image blending
