基于二维图像的三维重建技术研究

Research on 3D Reconstruction Technology Based on 2D Image

针对在拍摄过程中,三维图像转化为二维图像时容易丢失深度信息,使得通过该二维图像重建出所对应的三维图像比较困难的问题。依据双目视觉技术,采用两个摄相机来完成研究对象的三维重建,既减小了整个系统的成本,又简化了三维重建的操作过程。通过研究所搭建的双目平台相机拍摄图像以及进行相机标定,再根据改进的边缘提取算法结合Harris算法对图像的特征点进行检测,并利用SAD算法对图像的特征点进行立体匹配,最后利用三角测量原理提取到目标图像的深度信息,最终实现对矩形量块的图像的三维图像的重建。实验结果证明,该方法取得了满意的实验效果。

When taking a 2D image of a 3D scene,it is difficult to reconstruct the corresponding 3D image from the 2D image because of the loss of depth information.According to the binocular vision technology,two cameras are used to complete the 3D reconstruction of the research object,which not only reduces the cost of the whole system,but also simplifies the operation process of the 3D reconstruction.In this paper,the camera of the binocular vision platform is used to obtain the left and right camera images.After the camera calibration,the improved edge detection algorithm combined with the Harris algorithm is used to detect the feature points of the image.SAD algorithm is used for stereo matching of the feature points of the image.The depth information of the target image is extracted by using the triangulation principle.Finally,the 3D image reconstruction of the image of the rectangular measuring block is realized.This method has achieved satisfactory experimental results.