计算机工程与设计
計算機工程與設計
계산궤공정여설계
COMPUTER ENGINEERING AND DESIGN
2009年
14期
3422-3425
,共4页
彩色图像拼接%Harris角点检测算子%RANSAC算法%仿射变换%HSV变换
綵色圖像拼接%Harris角點檢測算子%RANSAC算法%倣射變換%HSV變換
채색도상병접%Harris각점검측산자%RANSAC산법%방사변환%HSV변환
color image mosaic%Harris comer detection operator,RANSAC algorithm%affine transform%HSV transform
针对彩色图像的拼接问题,提出了一种基于角点检测和仿射变换模型的图像配准算法.利用Harris算子在彩色图像的HSV空间的亮度分量V中检测角点,然后引入基于灰度相关交叉匹配和简化RANSAC算法的两级约束机制确定同名角点,并计算图像仿射变换模型参数,实现图像配准.同时结合人眼的视觉非线性给出了一种分段加权函数,通过对饱和度分量S和亮度分量V的处理,消除明显的拼接缝,实现色彩和亮度的平稳过渡.实验结果表明,该算法具有良好的拼接精度和拼接缝消除效果.
針對綵色圖像的拼接問題,提齣瞭一種基于角點檢測和倣射變換模型的圖像配準算法.利用Harris算子在綵色圖像的HSV空間的亮度分量V中檢測角點,然後引入基于灰度相關交扠匹配和簡化RANSAC算法的兩級約束機製確定同名角點,併計算圖像倣射變換模型參數,實現圖像配準.同時結閤人眼的視覺非線性給齣瞭一種分段加權函數,通過對飽和度分量S和亮度分量V的處理,消除明顯的拼接縫,實現色綵和亮度的平穩過渡.實驗結果錶明,該算法具有良好的拼接精度和拼接縫消除效果.
침대채색도상적병접문제,제출료일충기우각점검측화방사변환모형적도상배준산법.이용Harris산자재채색도상적HSV공간적량도분량V중검측각점,연후인입기우회도상관교차필배화간화RANSAC산법적량급약속궤제학정동명각점,병계산도상방사변환모형삼수,실현도상배준.동시결합인안적시각비선성급출료일충분단가권함수,통과대포화도분량S화량도분량V적처리,소제명현적병접봉,실현색채화량도적평은과도.실험결과표명,해산법구유량호적병접정도화병접봉소제효과.
An image registration method based on the comer detection and affine transform model is proposed. This method uses the Harris operator to detect the comers in the HSV space of the color images. A two-level restricted tactic based on cross match and simplified RANSAC algorithm to gain the homonymy corner points is introduced. A segment-weighted function based nonlinear characteristic of human vision is given to eliminate the mosaic line caused by the difference ofthe image color. The experiment indicates that this algorithm has good mosaic precision and nice fusion of pixel color.
