In allusion to the problem that the large object size is difficult to measure, this paper researches simple and quick measuring methods and presents a method of 3D measurement of object' s size based on image using digital camera. Firstly, we obtain images from multi-angles using a digital camera and detect characteristic points using Harris corner detection algorithm which is strong stability. Secondly, a new method of adaptive looking for calibration template based on the object to be tested in images is proposed in the processing of the calibration, and expands the range of application of the calibration target. Finally, we compute the world coordinates of the characteristic points to measure using projection matrix theory. Experiments prove that measuring results is within two percent, which show that this method can overcome the disadvantages of the traditional measurement and meet requirements of general industrial measurement and application scope is wide.%针对大型物体尺寸不易测量的问题,研究简单快速的测量方法;提出了一种利用数码相机完成物体尺寸的三维测量的方法;利用数码相机从多角度获取图像,采用稳定性强的Harris角点检测算法对图像进行特征点检测;在相机标定过程中提出了一种利用图像中被测物体自适应寻找标定模板的方法,扩大了标定模板的使用范围；最后利用投影矩阵理论计算特征点的世界坐标进行测量；实验证明测量结果误差在2.0％以内,说明该方法能够克服传统测量方法的缺点,满足一般工业的测量要求并且适用范围广.
展开▼
