基于图像法表征烟幕颗粒实验设计

Experimental design on smoke screen particle characterization based on image method

为表征烟幕颗粒大小和形状,开展基于图像方法的烟幕颗粒测量的实验研究。在现有的烟幕实验观测箱基础上,增加多点图像采集装置,采集烟幕颗粒图像。利用盲去卷积法、自适应阈值二值化、闭运算等图像处理方法消除颗粒图像离焦模糊、光照不均匀等不利影响;利用特征提取函数得到颗粒信息。通过多组实验验证,图像法适用于大空间烟幕颗粒表征,测量结果重复性好、可信度高,采用长轴特征以及长宽比特征表征烟幕颗粒更具合理性,但图像采集装置的布置方式会对最终测量结果产生影响,需合理布置。

In order to characterize the size and shape of smoke particles, the experimental study of smoke particle measurement based on image method is carried out. Based on the existing smoke experimental observation box, a multi-point image acquisition device is added to collect the smoke particle image. The image processing methods such as blind deconvolution, adaptive threshold binarization and close operation are used to eliminate the disadvantageous effects such as defocus blur and uneven illumination, and the particle information is obtained by feature extraction function. Through multiple groups of experiments, the image method is suitable for the characterization of smoke particles in large space, and the measurement results have good repeatability and high reliability, which is more reasonable to use the long axis feature and aspect ratio feature to characterize the smoke particles. However, the layout of the image acquisition device will have an impact on the final measurement results, so it needs to be reasonably arranged.