Substantial advances in quantitative infrared vision

红外热像仪是材料无损检测的理想工具。基本上，观测以被动或主动方式进行。被动模式包括利用材料内部产生的自然热量并观察温度分布。在主动模式下，使用外部热源在被研究的材料内产生热扰动。随后，利用红外成像仪分析了热波传播的异常，揭示了隐藏的缺陷。该方法成功的一个重要部分是基于热扰动（主动模式）如何传递到材料。另一个至关重要的因素是图像处理（主动和被动模式），因为它允许对观察到的现象进行解释。到目前为止，传统的热成像技术在检测和解释阶段都遇到了一些限制。拟议的研究计划侧重于研究解决关键问题的新方法。我们的短期目标将包括以下几个方面：(1)探索最佳主动热成像实验：目的是通过理论和实验分析不同空间和光谱变化的加热模式的影响；目的是提高缺陷的可检测性；(2)先进的处理方法：通过热成像信号重构（TSR）或主成分分析（PCA）等最新数据压缩算法求解逆问题的方法；计划中的创新例如神经网络（NN）；(3)主动热成像的热物理学：目标是可靠地测量热无损检测的各种新材料的热性能；热性能要求提供热模拟和缺陷表征代码。(4)低成本检测系统的设计：目标是设计低成本的红外热成像装置，可能被中小型加拿大公司使用，这些公司买不起高端设备。强大的内置处理与“微型”非冷却红外相机的新技术相结合，将导致这种系统的发展。

Infrared (IR) thermography is an ideal tool for the nondestructive testing of materials. Basically, the observation is carried out either in a passive or active mode. The passive mode consists in using the natural heat developed within the material and observing the temperature distribution as it appears. In the active mode, an external heat source is used to create a thermal perturbation within the material under investigation. Subsequently, an IR imager is used to analyze the thermal wave propagation whose anomalies reveal hidden flaws. An important part of the success of the method is based on how the thermal perturbation (active mode) is transmitted to the material. Another factor that is of paramount importance is image processing (active and passive modes) since it allows the interpretation of the observed phenomena. So far, conventional thermography techniques have encountered several limitations at both the detection and interpretation stages. The proposed research program focuses on investigating new ways of solving the key problems. Our short term objectives will cover the following directions: (1) Investigation of optimal active thermography experiments: the objective is to analyze via theory and experiments the effect of different spatially, and spectrally varying heating modes; the aim is to enhance the detectability of the flaws; (2) Advanced processing methods: methods for solving inverse problems through the state of the art in data compression algorithms such as thermographic signal reconstruction (TSR) or principle component analysis (PCA); planned innovations are for instance into neural networks (NN); (3) Thermophysics for active thermography: the goal is to reliably measure thermal properties of the vast range of new materials subjected to thermal nondestructive tests; thermal properties are required to feed thermal simulation and flaw characterization codes. (4) Design of low-cost inspection systems: the objective is to design low-cost IR thermography setups that might be used by small and medium-size Canadian companies which cannot afford high-end equipment. Powerful built-in processing combined with the new technology of "tiny" uncooled IR cameras will lead to the development of such systems.