摘要： 提出了一种将感知编组应用于平行四边形提取的算法。该算法利用逐级抽象感知的思想，在基于part 的形状表达的基础上，使用不确定性推理，达到了结构清晰简单、计算量小、抗噪能力强的效果。实验表明，该算法是一种有效的平行四边形提取算法。

关键词: 平行四边形提取；感知编组；不确定性推理；形状表达

Abstract: This paper proposes a parallelogram extraction algorithm applying perceptual organization. It uses perceptual organization as basic scheme, employing uncertainty reasoning and a shape expression form based on “part” to obtain simplicity, low computational overhead, and strong ability of anti-noise. Experimental results show it is an effective parallelogram extracting algorithm.

Key words: Parallelogram extraction; Perceptual organization; Uncertainty reasoning; Shape expression