采用了新近发展起来的边缘提取技术——细胞神经网络(CNN),从工业CT(Computed Tomography,计算机断层成像)体数据出发,提取被扫描工件的内外表面(称为边缘面)。当一个边缘面与沿某方向切片序列中的某切片重合时,采用二维边缘提取方法不易从该切片序列中提取出边缘面。针对这种情形,将工业CT体数据沿三个互相垂直的方向剖分,得到相应的切片序列。然后对每个切片,采用两组二维细胞神经网络实现边缘提取。再将同方向的切片边缘数据重组,得到对应方向的边缘体数据。最后,综合各方向的边缘体数据得到边缘面。该算法由于考虑了体数据点在三个方向的灰度变化,分割结果比仅考虑单一方向的算法更接近真实情形。对边缘分割后的体数据的三维显示表明,该算法能得到比较完整真实的边缘面。
To extract the internal and external surfaces(called edged surface) of the scanned workpiece from the industrial Computed Tomography(CT) volume data even the surface is included in a slice,edge extraction technology developed recently--Cellular Neural Network(CNN) is used.After getting the corresponding slice sequence along three mutually perpendicular directions,the edge extraction of a slice is realized by two sets of cellular neural network.Then the slice edge data is restructured to get the edge volume data along one direction,which will be integrated in three directions to gain the final edge surface.Because the gray change of a point along three directions are taken into account,this algorithm can get more complete edge surface than dealing with only one direction.Computer experiment results validate the validity of the algorithm.