东南大学学报(自然科学版)
東南大學學報(自然科學版)
동남대학학보(자연과학판)
JOURNAL OF SOUTHEAST UNIVERSITY
2013年
z2期
283-285
,共3页
图像匹配%核线%同名像点
圖像匹配%覈線%同名像點
도상필배%핵선%동명상점
image matching%epipolar line%homologous image points
为了提高图像匹配的效率,在同名核线上搜索同名像点时,研究缩小同名像点在同名核线上搜索范围的方法,提高匹配的速度和精度。首先根据研究区域的地形起伏情况确定地面点高程可能的取值,并求出左图像对应的地面点的点投影系数的取值范围;然后求出地面点对应的高低2个点三维坐标;最后根据高低点的三维坐标求出右图像上同名像点的范围,确定核线上局部搜索区间。通过交通视频检测立体图像的实验,给定2m的地形起伏,同名像点在核线上的搜索区间大大缩小。该方法提高了匹配速度和匹配成功率。在计算机视觉和数字摄影测量领域具有广泛的应用。
為瞭提高圖像匹配的效率,在同名覈線上搜索同名像點時,研究縮小同名像點在同名覈線上搜索範圍的方法,提高匹配的速度和精度。首先根據研究區域的地形起伏情況確定地麵點高程可能的取值,併求齣左圖像對應的地麵點的點投影繫數的取值範圍;然後求齣地麵點對應的高低2箇點三維坐標;最後根據高低點的三維坐標求齣右圖像上同名像點的範圍,確定覈線上跼部搜索區間。通過交通視頻檢測立體圖像的實驗,給定2m的地形起伏,同名像點在覈線上的搜索區間大大縮小。該方法提高瞭匹配速度和匹配成功率。在計算機視覺和數字攝影測量領域具有廣汎的應用。
위료제고도상필배적효솔,재동명핵선상수색동명상점시,연구축소동명상점재동명핵선상수색범위적방법,제고필배적속도화정도。수선근거연구구역적지형기복정황학정지면점고정가능적취치,병구출좌도상대응적지면점적점투영계수적취치범위;연후구출지면점대응적고저2개점삼유좌표;최후근거고저점적삼유좌표구출우도상상동명상점적범위,학정핵선상국부수색구간。통과교통시빈검측입체도상적실험,급정2m적지형기복,동명상점재핵선상적수색구간대대축소。해방법제고료필배속도화필배성공솔。재계산궤시각화수자섭영측량영역구유엄범적응용。
In order to improve the efficiency of image matching, this paper proposes a method of narrowing the searching scope while searching the corresponding image points on the corresponding epipolar line, thus improving the matching speed and precision.First, the possible values of ground elevation are determined according to the terrain of the research area, and the point projection coeffi-cient range of the terrain point corresponding to the left image is calculated.Secondly, the three-di-mensional coordinates of the terrain point corresponding to the high-low points are calculated.Final-ly, according to the three-dimensional coordinates of the high-low points, the range of the corre-sponding image points on the right image is found and the searching scope of the epipolar line is de-termined.Given a two-meter accidented terrain, the searching scope of the corresponding image points on the corresponding epipolar line is greatly narrowed through the experiment of traffic video detection stereo image.It proves that the proposed method can improve the matching speed and the success rate, and it is widely used in the field of computer vision and digital photogrammetry.