CAJ | 학술논문

提出了基于近似最小距离场提取二值图像的8-连通骨架的算法。该算法对图像中的每个像素根据其与边界的相对距离进行整数编码，形成近似最小距离场，将该距离场中的几何邻接的、具有局部最大值的像素形成聚类，对聚类进行细化，用最短路径将不同的细化后的聚类连接起来。该算法简单，将其在实验数据集上进行实验，结果证明算法具有很高的效率。
제출료기우근사최소거리장제취이치도상적8-련통골가적산법。해산법대도상중적매개상소근거기여변계적상대거리진행정수편마，형성근사최소거리장，장해거리장중적궤하린접적、구유국부최대치적상소형성취류，대취류진행세화，용최단로경장불동적세화후적취류련접기래。해산법간단，장기재실험수거집상진행실험，결과증명산법구유흔고적효솔。
This paper proposes an algorithm for extracting 8-connected skeletons of 2D binary images. Each interior pixel in the 2D image is encoded with an integer code according to its relative distance from the object border to form an approximate minimum distance field. Cluster is defined as a set of geometrically connected local maximum pixels with the same distance value. And all the clusters are thinned, and connected with the shortest paths. The proposed algorithm is simple, and the results acquired by the algorithm on an experimental data demonstrate its efficiency.