ORB feature detection-based three-dimensional reconstruction method

The invention discloses an ORB feature detection-based three-dimensional reconstruction method, and belongs to the technical field of feature detection. The method comprises the steps of obtaining a real-time image through a binocular camera mounted on an unmanned aerial vehicle support platform, and performing preprocessing; calibrating the camera; working out internal and external parameters anda distortion coefficient of the camera to correct pixel coordinates; detecting feature points of the image by using ORB feature detection; performing matching by using an FLANN; working out spatial coordinates of the feature points; and performing three-dimensional modeling on spatial discrete points by utilizing OpenGL. The method can increase system execution speed, has rotational invariance and anti-noise interference robustness at the same time, and greatly improves the precision of real-time three-dimensional reconstruction. 本发明公开了种基于ORB特征检测的三维重建方法，属特征检测技术领域，通过安装在无人机云台上的双目摄像头获取实时图像及预处理，对摄像机进行标定，求解出相机内外参数和畸变系数对像素