基于SVM的填充轮廓分类方法

Filling Contour Classification Method Based on SVM

为提高3D打印技术填充过程中填充路径对填充轮廓几何特征的适应性,提出一种基于SVM的多边形轮廓分类方法。分析与填充轮廓相关的可测变量多边形的圆度、面积/周长比、锐角占比;利用机器学习方法建立SVM模型,对多边形类型进行分类预测。该方法可以避免逐一分析复杂的几何学参数,并且可高效、准确地对待填充轮廓进行自适应路径选择。结果表明:利用该方法可以取得良好的分类效果,模型预测精度达到90%以上,基本满足实际加工要求。

In order to improve the adaptability of the filling path to the geometric features of the filled contour during the 3D print⁃ing technology,a polygon contour classification method based on SVM was proposed.The measurable variables associated with the fill⁃ing profile were analyzed,such as polygon degree,area/perimeter ratio,acute angle account for the percentage of all angles;SVM model was established by using machine learning method to classify and predict polygon types.By using this method,the analysis of complex geometric parameters one by one can be avoided,and the adaptive path for the filled contour can be efficiently and accurately selected.The results show that by using this method,a good classification effect can be achieved,and the prediction accuracy of the model is more than 90%,which basically meets the actual processing requirements.