基于表面测温的多区域交界面形状的识别研究是红外无损检测技术从定性到定量发展的理论基础。针对多区域交界面形状的常用识别算法受测量误差和测点数目影响相对较大或算法构造相对困难的问题，在改进的一维修正算法和分散模糊推理算法的基础上，提出识别多区域交界面形状的一维加权法，并对两种材料的管壁交界面形状的识别问题进行了研究。讨论了管壁交界面形状初始假设、检测表面测量误差以及测点数目等因素对一维加权法识别结果的影响。数值算例证明了一维加权法的有效性和高效性。

The identification of interface geometry of a multiple region domain based on surface temperature measurement is the theoretical basis for the development of thermographic non-destructive inspection from qualitative assessment to quantification. The commonly used algorithms to identify the interface geometry of the multiple region domain are greatly affected by measurement errors and number of measuring points or are relatively difficult to construct. Concerning these problems, one-dimensional weighting method was proposed to identify the interface geometry of the multiple region domain based on the modified one-dimensional correction method and decentralized fuzzy inference method, subsequently, inverse identification of interface geometry of pipe wall of two kinds of materials was also studied. The influence of the initial interface geometry guess, measurement error and number of measuring points at the inspection boundary on the identification result of one-dimensional weighting method were studied. Lastly, numerical examples were given to demonstrate the effectiveness and efficiency of the one-dimensional weighting method.