六角网格中三维物体模式识别算法

提出一些算法,这些算法预期成为在六角网格中实现对三维物体快速、并行的模式识别的整体方案的一部分。这些算法包括在7-像素六角邻域之内基于灰度差值的边缘检测、局部边缘减薄和基于7-点六角掩模的特征提取。通过串行计算机模拟,这些算法对诸如方体等一些简单物体的真实六角网格数字图像的实验性验证已经显示了在快速识别三维物体轮廓的关键性特征方面的成功。

Algorithms for Pattern Recognition of 3-D Objects on a Hexagonal Grid

In this contribution,we bring up several algorithms,which are expected to become part of an overall scheme for achieving rapid pattern recognition of 3-D objects based on parallel computation architecture on a hexagonal pixel grid. These algorithms include edge detection using gray-level differences within a seven-pixel hexagonal neighborhood, local edge thinning, and feature extraction using seven-point hexagonal masks. Experimental tests of these algorithms by serial computer simulation on real hexagonal-array digital images of some simple objects such as a cube have shown success at rapidly recognizing essential features of the objects skeletons.