基于阈值分割与多边形扫描求交的明暗恢复形状重构模型冗余信息去除

Redundant Information Removing of SFS Reconstructed Model Based on Threshold Segmentation and Polygons Intersecting

针对明暗恢复形状重构模型存在图像背景信息问题，提出基于阈值分割与多边形扫描求交的冗余信息去除方法。通过阈值分割将图像转换为二值图像，采用形态学方法修补二值图像中的缺陷，并利用微分算子提取图像的二维轮廓；通过多边形扫描求交将图像分割为目标区域和背景区域；根据由阴影恢复形状原理重构三维模型，并通过二维轮廓与三维形貌的信息匹配去除图像背景冗余信息。实例表明，该方法能够有效去除明暗恢复形状重构模型中的图像背景冗余信息，保证重构模型的形状精度。

To remove the redundant points of shape from shading reconstructed model, a practical methods based on threshold segmentation and polygons intersecting is proposed. The image is converted into binary image through threshold segmentation technique, the defects contained in image are remedied by morphological method, and the 2D contour is extracted by using the differential operator, the binary image is divided into objective region and background region by using the polygons intersecting method; the 3D model is reconstructed by using shape from shading method, and the redundant points of shape from shading reconstructed model are removed through fusing the information of 3D profile and 2D contour. Experiment results showed that the method proposed can effectively remove redundant points of shape from shading reconstructed model, and it is helpful for improving the shape precision of shape from shading reconstructed model.