【摘要】 研究从全息重建颗粒图像中进行颗粒识别和定位的方法.利用小波函数重建颗粒全息图,采用灰度阈值自动判定方法对所重建的三维颗粒场图像进行颗粒图像与背景的分离和颗粒判定.根据重建颗粒图像灰度的空间分布特点,采用灰度和颗粒面积双判据方法实现颗粒空间位置,特别是流场深度方向的准确定位.对已知记录距离的理想模拟颗粒群全息图进行重建测试,同时在共轴全息试验台上对已知记录距离的颗粒试验片进行颗粒识别和定位算法的测试和验证.结果表明,所提出的灰度阈值自动判定方法准确、有效,最大灰度和颗粒面积双判据方法能够更加准确地进行颗粒空间定位.更多还原

【Abstract】 Particle identification and location measurement from the reconstructed holographic particle images were investigated.Based on the reconstruction of particle hologram by wavelet transform,a self-adapting grayscale threshold was presented to extract the particle information from the reconstructed images.According to the spatial distribution characteristics of the grayscale of the reconstructed particle images,two parameters,grayscale and area of particle image distribution in optical axis direction,were used to correctly determine the 3D location,especially the depth position of the particles.A simulated particle hologram as well as real holograms from an experiment using a particle slide with known recording distance on a digital in-line holographic test bench was tested.The results show that the self-adapting grayscale threshold is accurate and effective in particle identification,and the double-criterion methods are more accurate in particle location than the single-criterion method.更多还原