基于图像处理的微小孔径测量系统设计
Design of Micro-Aperture Measurement System Based on Image Processing

随着现代科技和工业的迅速发展，高精度和微型化工艺已成为重要的发展趋势。然而对于高精度测量，传统的测量技术面临着测量数据少、精度不够和存在人工误差等问题，因此设计一个完整的计算机辅助测量系统具有重要的意义。本文提出的微小孔径测量系统包括三个部分：图像采集、图像处理以及孔径测量，其中，图像处理是关键的一步，图像处理的效果与测量结果密切相关。图像处理部分首先采用全变分模型对盲孔图像进行去噪，然后用基于SURF的算法进行图像配准，针对图像融合，本文提出了一种基于改进的边缘保持滤波与脉冲耦合神经网络的多聚焦图像融合算法，采用Canny-Zernike矩亚像素边缘检测方法对盲孔图像做边缘检测，再对得到的边缘图像进一步做形态学滤波，最后采用最小二乘拟合法对盲孔边缘进行拟合即可得到盲孔直径。实验结果表明，本文设计的测量系统得到的测量误差较小，测量结果比较稳定。
With the rapid development of modern technology and industry, high precision and miniaturization process has become an important development trend. However, for high-precision measurement, the traditional measurement technology is faced with the problems of less measurement data, insufficient accuracy and artificial error. Therefore, it is of great significance to design a complete computer-aided measurement system. The micro-aperture measurement sys-tem proposed in this paper includes three parts: image acquisition, image processing and aperture measurement. Among them, image processing is a key step, and the effect of image processing is closely related to the measurement results. The image processing part first uses the total variation model to denoise the blind hole image, and then uses the SURF-based algorithm for image registration. For image fusion, this paper proposes a multi-focus image fusion algorithm based on improved edge-preserving filtering and pulse-coupled neural network. The Canny-Zernike moment sub-pixel edge detection method is used to perform edge detection on the blind hole image, and then obtained edge image is further subjected to morphological filter-ing. Finally, the least squares fitting method is used to fit the edge of the blind hole to obtain the diameter of the blind hole. The experimental results show that the measurement error obtained by the measurement system designed in this paper is small, and the measurement result is relatively stable.