Simple correction methods of infrared thermography for building exterior surfaces

Abstract

There is a need to improve the accuracy of infrared thermography for measuring the temperature distribution of building exterior surfaces. Thermography is useful for building ventilation and thermal analysis, as well as understanding city ventilation and the urban heat island phenomenon. The key in correcting infrared images is to quantify accurately the reflected infrared contribution of surrounding surfaces as well as that of the atmosphere. Two new methods are proposed here for correcting measured temperature distribution of building exterior surfaces by infrared thermography. The first is a simplified version of the aluminium mirror method first developed by Datcu et al (2005). The second is a temperature-based method using the measured surface temperatures at one or more points on the same surface. We present two series of specially designed experiments to evaluate the two proposed methods and two other commonly used/default methods. The temperature-based method can be used effectively to correct the infrared thermography results for measured surfaces with both specular and diffuse reflection, while the aluminium mirror method can only be used for surfaces with diffuse reflection. A good agreement was observed between the corrected infrared thermography results using the temperature-based method and thermocouple measurements with a mean error ranging between 0.07-0.36°C for concrete surfaces, 0.05-0.11°C for glazing surfaces and 0.17-0.41°C for tile surfaces. The centre location of the target surface can be used as an ideal sampling point for correcting the entire measured surface. The proposed new methods were further verified on a real building external surface. The method is also applicable to measurement of indoor surfaces.