由粗到细的图像浅浮雕生成算法

Course-to-fine Bas-relief Generation Algorithm from Images

针对现有基于三维场景的浅浮雕生成算法输入要求高等问题,提出了一种基于图像的细节增强的浅浮雕生成算法.首先由图像的灰度值提取高度场,通过对高度场的光滑处理得到浮雕的大致轮廓.然后对图像进行细节增强处理,提取图像的细节部分.最后将细节部分转化为高度场,并与上述光滑浮雕相互叠加,得到细节增强的浅浮雕模型.该算法的输入要求低,简单易于实现.实验表明算法能够利用图像生成细节增强的浅浮雕,具有实用价值.

Current methods for generating digital bas-relief based on 3D scenes have high input requirements, this paper proposes a detail enhancement bas-relief generation algorithm based on the image. By the method, first we extract the height field based on the gray values of the image, and a rough shape of the bas-relief can be obtained by smoothing the height field. Then we apply some detail enhancement operations on the image and detect the details of the image. Finally, we turn the detail part into a detail height field and add it to the smooth height field to generate a detail enhancement bas-relief model. The approach has low input requirements as well as easy to implement for its simpleness. Experimental results show that the approach can be applied to generate detail enhancement bas-reliefs, which has practical value.