针对太赫兹光场成像中由前景遮挡物影响感兴趣信息采集的问题,提出基于多视角合成孔径的太赫兹光场成像去遮挡算法。在分析太赫兹光场成像数字重聚焦原理的基础上,首先使用太赫兹焦平面阵列相机来采集太赫兹光场的原始数据,然后通过确定最小泛化误差来定位数字重聚焦深度,最后应用经验模式分解(EMD)法对重构图像进行增强处理,得到目标物轮廓分明、抑制遮挡物干扰的太赫兹图像。实验结果表明,太赫兹光场技术与合成孔径技术的结合可有效降低遮挡物的影响,另外证实了EMD法具有改善太赫兹图像质量的能力。

Spray behavior is regarded as one of the main factors influencing engine performances, fuel consumption and emissions for diesel engines. Under high injection pressure, diesel spray behaviors are extremely sensitive to the nozzle internal geometries, especially the geometric structures of orifice entrance. Based on the synchrotron radiation X-ray tomography technique, the 3D digital models of nozzle tips can be constructed. A new automatic method is presented to reveal the inlet structures according to these nozzle orifice models. The planes passing through the orifice axis are determined and used to cut the models, and then the corresponding cutting images are applied to measure the inlet chamfer radii around the orifice axis automatically. The orifices of a single-hole nozzle and an eight-hole nozzle are measured according to this method. The results show that this method can automatically measure the orifice inlet chamfer radii around the orifice axis with high precision. The obtained inlet chamfer radius shows the whole profile of the orifice entrance, which is a precise feedback for nozzle designing and manufacturing, and it also provides precise geometrical boundary conditions for the study of spray behaviors.