【摘要】 工程图纸矢量化之前 ,必须对图纸进行预处理 .本文提出了根据具有不同的 L ipschitz指数的信号和噪声的局部模极大值具有不同的传播特性的理论 ,应用小波变换模极大值去除扫描图纸的噪声的方法 ,并给出了具体算法 .同时充分利用原图纸的空域上的线条有别于污点的特征 ,对经去噪后的图纸采用逻辑平均抽取算法 .去除大块的污点 .实验结果证明了该算法的可行性更多还原

【Abstract】 In This paper, based on propagation properties of local modulus maximum of the signal and noise which have different Lipschitz exponents , a method of eliminating noise by using wavelet modulus maximum is proposed. At the same time we designed a logical average method to deal with the big dirty in engineering blueprint using the characteristic that the line is different from the big dirty.The experimental results show that the algorithm is feasible.更多还原