CAJ | 학술논문

骨架是形状表示的重要特征,传统的骨架算法往往不能直接用于物体识别,且连通性难以保证.用求向量内积的方法对骨架进行提取,通过距离变换得到连接图像各点与最近边沿点的向量,并利用内积计算求取两个相邻点向量的内积值;再根据内积值进行骨架种子点的选择,经两次骨架生长处理得到连通的骨架.实验证明本算法复杂度低,能很好保证骨架的连通性.
골가시형상표시적중요특정,전통적골가산법왕왕불능직접용우물체식별,차련통성난이보증.용구향량내적적방법대골가진행제취,통과거리변환득도련접도상각점여최근변연점적향량,병이용내적계산구취량개상린점향량적내적치;재근거내적치진행골가충자점적선택,경량차골가생장처리득도련통적골가.실험증명본산법복잡도저,능흔호보증골가적련통성.
The skeleton is an important feature in the representation of shapes.Traditional skeletonization algorithm can not be used for skeleton recognition directly,and the connectivity property of the skeleton is not guaranteed.A skeleton extraction algorithm is proposed based upon vector inner-product.The vectors connecting image points to the nearest border points are determined by distance transform,the innerproduct is calculated between vectors from neighboring points.The seeds of skeleton are selected by the value of inner-product.A well connected skeleton is determined by two steps of skeleton growth.The proposed algorithm is proved to be with low time complexity.The skeleton is produced to be well connected.