论文部分内容阅读
准确提取竹材横端面的完整轮廓对竹材加工自动化有着至关重要的作用。以聚类理论为基础，选用符合竹材横端面颜色特性的Lab颜色空间，提出一种基于聚类理论的竹材图像分割算法。该算法采用Lab颜色空间a、b通道组成适合竹材图像的特征向量，利用kmean聚类算法对组合特征向量进行分类实现对竹材横端面图像的分割，并利用圆拟合算法实现对竹材端面的最终框定。试验结果表明此方法能够对竹材横端面图像进行分割，并能够有效地框定出竹材端面，算法在时间性能上能够满足后续的处理要求。“,”Extracting complete contour from bamboo image accurately plays an important role in bamboo processing automation. Based on the clustering theory and bamboo cross-section color features in the Lab color space, a novel bamboo image segmentation algorithm is presented in this paper. The feature vector is extracted from a and b channel of Lab color space in proposed algorithm. At last we use the k-mean clustering algorithm to classify feature vector and implement bamboo cross-section image segmen-tation, a circle fitting method is used to the bamboo frame in the end. Experimental results showed that the proposed algorithm can be effectively applied to the bamboo cross-section image segmentation and frame. The algorithm can meet the requirement of follow-up processing in time performance.