逆向工程中一种新的特征识别算法

从某种角度上说,逆向工程是从已有实物的测量数据点中提取其实体特征再进行模型重建的过程。本文提出了一种新的特征识别算法,其首先采用基于面积和法矢准则的数据分割技术,对测量数据点进行数据分割。然后从特征所包含的分割面(简称特征分割面)中提取能够惟一标识该特征的4种特征编码,分别为:表述特征截面形状的截面编码、描述特征凹凸性的凹凸编码、显示特征二维俯视轮廓形状的轮廓编码以及反映特征二维俯视轮廓是否封闭的开闭编码。最后将这4种编码输入到基于人工神经(BP)网络的自动特征识别系统中,识别出特征类型并提取特征参数,从而实现特征重建。着重研究并实现了从特征分割面中提取特征编码的算法,并验证了算法的有效性。

New Algorithm for Feature Recognition in Reverse Engineering

In a sense,reverse engineering is the process of extracting the solid features from the measured point data of the existing parts.A novel algorithm is proposed for feature recognition.Firstly,the measured point data are segmented by the segmentation technology based on area and normal rules;and then four feature codes are acquired from its segments to illustrate the affiliated feature,such as section codes indicating the section figure of the feature model,concave-convex code describing whether a selected profile is convex or concave,profile code representing the 2-D geometric shape of a feature profile and open/close code denoting whether the 2-D geometric profile is open or closed.Finally,the acquired codes are input to the automatic system of feature recognition based on the BP network,thus recognizing the corresponding feature type and achieving its inclusive parameters.The new method for acquiring these codes from feature segments is investigated with emphasis and is proved to be effective and flexible by experiments.