基于自适应匹配的矢量图符号模糊识别仿真

Simulation of Fuzzy Recognition of Vector Symbols Based on Adaptive Matching

针对传统矢量图符号识别方法易出现符号匹配误差,导致识别精度较低的问题,提出一种矢量图符号模糊识别方法。采用自适应匹配算法,结合背景回波提取符号特征点,并将初始表达空间映射至低维空间内,实现符号特征库空间维数的压缩,同时保留原本信息。将已知特征待分类符号样本与模板库内模板匹配,并获取样本与模板间的距离,依据匹配交互学习符号样列组建符号库,通过符号样本坐标在复杂图纸内搜索与之相同或类似符号,实现对矢量图符号的模糊识别。仿真结果证明,所提方法能够精确识别出矢量图符号,与传统方法相比应用效果更理想。

Aiming at the problem of low recognition accuracy caused by symbol matching error in traditional vector graph symbol recognition method, a fuzzy recognition method of vector graph symbol is proposed. The adaptive matching algorithm and background echo were applied to extract the feature points of symbols. The low dimensional space accepted the mapping from the initial representation space. Meanwhile, the original information was preserved. The symbol samples with known features were matched with the template in the template library for getting the distance between the sample and the template. The symbol library was established via matching the sample columns of interactive learning symbols. The same or similar symbols were established based on the coordinates of symbol samples, thus completing the fuzzy recognition of vector symbols. The simulation results show that the recognition effect of this method is superior to that of traditional methods, and the symbol of vector graph can be accurately recognized.