EAGER: Laser 4D light field thermoreflectance (TR) imaging for non-intrusive high accuracy temperature measurement of 3D targets with high spatial and high temporal resolutions

首席研究人员将开发一种技术来确定三维(3D)表面或物体上的温度分布。这项技术类似于3D摄影，只是在绘制温度地图方面增加了更多、更复杂的挑战。一台带有微透镜阵列的相机将对准物体。因为这些镜头以不同的角度投射图像，所以达到了三维的效果。同时，由于表面各点反射光的强度与温度有关，因此可以确定三维表面上的温度。除了指导研究生和本科生外，PI还将涉及高中教师，特别是来自少数民族人口较多的学校的教师。拟议的研究将创建一个用于热成像的微透镜阵列(MLA)，用于提取热信息的数值算法，以及确定LED封装在运行条件和瞬变状态下的瞬时3D温度分布。作者将应用一种称为4D光场成像的方法，该方法能够通过透镜阵列对3D表面进行成像，每个透镜对对象具有不同的视角，以将目标的图像(即全光图像)投影到CCD传感器上。然后，使用PI先前开发的算法，可以从多个视角重建表面的3D图像。基于表面随温度变化的反射率和反射光强度(即它们的热反射率)，可以识别温度变化。该技术可以获得高精度(~10mK)、高空间分辨率(~一半光波长，即衍射受限)和纳秒级的时间分辨率。因此，该技术可用于稳定、准稳定和瞬变状态的热成像，实现了包括极端条件、发光二极管(LED)和激光二极管(LD)在内的广泛应用。

The principal investigator will develop a technique to determine the temperature distribution on a three-dimensional (3D) surface or object. The technique is similar to 3D photography, with the added and much more complicated challenge in mapping out the temperature. A camera with an array of micro-sized lenses will be aimed at the object. Because at different angles these lenses project the images, the effect of three dimensions is achieved. At the same time the intensities of the light reflected from various points of the surface is related to temperatures, the temperature on the 3D surface can be determined. The PI will also involve high school teachers, especially those from schools with high minority population, in addition to supervising graduate and undergraduate students.The proposed study will create a microlens array (MLA) for thermal imaging, numerical algorithms to extract thermal information, and determination of transient 3D temperature profiles of LED packages during operating conditions and in transient state. The author will apply an approach known as 4D light field imaging, which enables imaging of 3D surfaces by the array of lenses, each having a different perspective on the object, to project images of the target (i.e., plenoptic images) onto a CCD sensor. The 3D image of the surfaces can then be reconstructed from the multiple perspectives, using an algorithm previously developed by the PI. Based on the temperature-dependent reflectivity and the reflected light intensity of the surfaces (i.e., their thermoreflectance), temperature changes can be identified. The proposed technique can achieve high accuracy (~ 10 mK), high spatial resolutions (~ half of light wavelength; i.e. diffraction-limited), and nano-second temporal resolutions. The technique can therefore be used for steady, quasi-steady, and transient state thermal imaging, enabling a wide range of applications including extreme conditions, light emitting diodes (LEDs) and laser diodes (LDs).