Wang Zhengzhou, Hu Bingliang, Yin Qinye
School of Electronic & Information Engineering, Xi'an Jiaotong University, Xi'an, China.
University of Chinese Academy of Sciences, Beijing, China.
PLoS One. 2017 Feb 16;12(2):e0171415. doi: 10.1371/journal.pone.0171415. eCollection 2017.
The schlieren method of measuring far-field focal spots offers many advantages at the Shenguang III laser facility such as low cost and automatic laser-path collimation. However, current methods of far-field focal spot measurement often suffer from low precision and efficiency when the final focal spot is merged manually, thereby reducing the accuracy of reconstruction. In this paper, we introduce an improved schlieren method to construct the high dynamic-range image of far-field focal spots and improve the reconstruction accuracy and efficiency. First, a detection method based on weak light beam sampling and magnification imaging was designed; images of the main and side lobes of the focused laser irradiance in the far field were obtained using two scientific CCD cameras. Second, using a self-correlation template matching algorithm, a circle the same size as the schlieren ball was dug from the main lobe cutting image and used to change the relative region of the main lobe cutting image within a 100×100 pixel region. The position that had the largest correlation coefficient between the side lobe cutting image and the main lobe cutting image when a circle was dug was identified as the best matching point. Finally, the least squares method was used to fit the center of the side lobe schlieren small ball, and the error was less than 1 pixel. The experimental results show that this method enables the accurate, high-dynamic-range measurement of a far-field focal spot and automatic image reconstruction. Because the best matching point is obtained through image processing rather than traditional reconstruction methods based on manual splicing, this method is less sensitive to the efficiency of focal-spot reconstruction and thus offers better experimental precision.
在神光Ⅲ激光装置上,纹影法测量远场焦斑具有成本低、激光光路自动准直等诸多优点。然而,当前远场焦斑测量方法在手动拼接最终焦斑时,往往存在精度和效率较低的问题,从而降低了重建的准确性。在本文中,我们引入一种改进的纹影法来构建远场焦斑的高动态范围图像,并提高重建精度和效率。首先,设计了一种基于弱光束采样和放大成像的检测方法;使用两台科学级电荷耦合器件(CCD)相机获取聚焦激光辐照度远场主瓣和旁瓣的图像。其次,利用自相关模板匹配算法,从主瓣切割图像中挖出一个与纹影球大小相同的圆,并用于在100×100像素区域内改变主瓣切割图像的相对区域。挖出圆时旁瓣切割图像与主瓣切割图像之间相关系数最大的位置被确定为最佳匹配点。最后,采用最小二乘法拟合旁瓣纹影小球的中心,误差小于1像素。实验结果表明,该方法能够实现远场焦斑的准确、高动态范围测量及自动图像重建。由于最佳匹配点是通过图像处理而非基于手动拼接的传统重建方法获得的,因此该方法对焦斑重建效率的敏感度较低,从而具有更好的实验精度。