中文摘要：

提出一种利用编码和极线约束相结合实现点匹配的方法，该方法利用圆环式编码点作为识别码标识区域。此类编码点具有旋转、缩放、变形的无关性，测量点是与编码点中心圆斑大小一样的圆斑。针对大尺寸测量对象中点的匹配问题，本文提出了分区的思想。首先进行区域匹配，即粗匹配；然后利用极线约束匹配小区域测量点，即细匹配。这样就完成了整个匹配。试验表明该方法能够有效地降低数据处理时间。提高匹配率和自动化程度。

英文摘要：

A point matching method with coded points and epipolar constrain is described. The circular ceded points are known as identifiers that are used to identify distracts. Identifying the ceded points is not affected when the coded points come forth rotation, zooming and deformation. The radius of the measured point is the same as the radius of the ceded points circle. To solve the point matching in huge shape measurement, the idea of the plotting area is presented. First, the subareas are matched, namely, crassitude matching, then the measured points are matched with epipolar constraint in the subareas of left and right images. The resuh shows that this method is valid and reliable in lowering data processing time, improving image matching veracity and au- tomatic degree.