CH-24-C108 - Utilization of Contrast-to-Noise Ratio for Identifying Thermal Anomalies in Northern Buildings with a Low-Cost IR Camera

It is becoming increasingly common to carry out thermal diagnostics on existing buildings, with the aim of assessing heat loss locations in the building and prioritizing where retrofit is necessary. These energy audits employ various non-destructive techniques, including infrared thermography (IR). Given the unique challenges faced by indigenous communities in northern Canada, the overall aim of the study is to evaluate the feasibility of using a low-cost IR camera for detecting thermal anomalies on northern buildings. To accomplish this, thermal anomalies were captured using both a low-cost IR camera and a higher-performance IR camera inten different northern buildings located in Quaqtaq, Nunavik, Canada. The Contrast-to-Noise Ratio (CNR) calculation is employed to assess the performance of the IR cameras and the severity of thermal anomalies. In this paper, we conduct a sensitivity analysis regarding the selection of the anomaly surface and the surface around the anomaly on thermal images for the calculation of the ratio. This analysis develops a method for selecting thermal surfaces to compute normalized CNR for all images showing a thermal anomaly. Furthermore, the study established that a thermal anomaly is identifiable when the calculated CNR is greater than 4. Lastly, IR images showing air infiltration obtained the highest CNR, indicating the severity of this anomaly, while thermal bridges obtained the lowest CNR values.