Abstract: Since there are abundant lines feature，plane feature in 3D space，utilizing these information to reconstruct 3D objects has a very broad prospects.However，due to the negative influence of noise and algorithm of extracting straight line，concavo-convex critical edging lines can not be extracted.Thus，how to recognize those not extracted concavo-convex critical edging lines leaves much to be desire.This paper suggests a novel algorithm to resolve this problem by following：Firstly，utilizing matched corresponding lines does space intersection to get space lines.Secondly，utilizing space lines fits space planes by parameter adjustment with constraint.Thirdly，space planes intersect to get concavo-convex critical edging lines.Finally，experimental results prove that our recognition algorithm has correctness，precise and practicality.

摘要： 由于三维空间中存在大量的线、面特征，所以基于线、面特征进行三维重建具有十分广阔的前景。然而，由于噪声和提取直线算法的影响，凹凸面的关键边界直线未能提取出来，那么如何识别无法提取出的关键凹凸面边界直线，是一个有待解决的问题。提出了通过同名直线（对应直线）前方交会出空间直线、空间直线基于带限制条件的间接平差拟合出空间平面、空间平面再相交，交出关键边界直线的新算法，解决了上述关键边界直线缺失的问题。通过实验，验证了该识别算法的正确性、有效性和实用性。