摘要
针对柔性印刷电路板(FPC)生产检测过程中的效率、精度等要求,提出一种基于模板与离散余弦变换(DCT)描述子的配准方法。首先,为了解决配准的精度问题,由电路板文件工程图解析得到对应图像,为后续处理提供更好的模板图,再利用离散余弦描述子的频域能量聚集特点,在模板与被检测图像之间建立对应关系,从而计算出对齐参数。其次,使用mipmap技术处理模板,为配准阶段提供不同精度的模板图片,从而有助于精度及效率的提升。实验结果表明:该改进算法的最大误差和平均误差均比原方法小;对于电路板这种大分辨率的图像,在mipmap采样的模板条件下,可以获得较高的精度,且处理速度有较大的提升。
Aiming at the requirements of efficiency and accuracy in the process of flexible printed circuit board(FPC)production detection,a registration method based on template and discrete cosine transform(DCT)descriptor is proposed.First,in order to solve the problem of registration accuracy,the corresponding image is obtained by analyzing the engineering drawing of circuit board file,which provides a better template for subsequent processing.Then,the corresponding relationship between the template and the detected image is established by using the frequency-domain energy aggregation characteristics of discrete cosine descriptor,so as to calculate the alignment parameters.Secondly,mipmap technology is used to process the template,which provides different precision template images for the registration stage,which is helpful to improve the accuracy and efficiency.The experimental results show that:the maximum error and average error of the improved algorithm are smaller than the original method;for the circuit board such a large resolution image,under the condition of mipmap sampling template,it can obtain higher accuracy,and the processing speed is greatly improved.
作者
王帆
伍世虔
WANG Fan;WU Shi-qian(School of Computer Science and Technology,Wuhan University of Science and Technology,Wuhan 430065,China;School of Information Science and Engineering,Wuhan University of Science and Technology,Wuhan 430081,China)
出处
《计算机与现代化》
2021年第10期57-62,共6页
Computer and Modernization
基金
国家自然科学基金资助项目(61775172)。
