Developing a semi/automated protocol to post-process large volume, High-resolution airborne thermal infrared (TIR) imagery for urban waste heat mapping
Date
2014-09-30
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In collaboration with The City of Calgary 2011 Sustainability Direction and as part of the HEAT (Heat Energy Assessment Technologies) project, the focus of this research is to develop a semi/automated ‘protocol’ to post-process large volumes of high-resolution (H-res) airborne thermal infrared (TIR) imagery to enable accurate urban waste heat mapping. HEAT is a free GeoWeb service, designed to help Calgary residents improve their home energy efficiency by visualizing the amount and location of waste heat leaving their homes and communities, as easily as clicking on their house in Google Maps. HEAT metrics are derived from 43 flight lines of TABI-1800 (Thermal Airborne Broadband Imager) data acquired on May 13-14, 2012 at night (11:00 pm - 5:00 am) over The City of Calgary, Alberta (~825 km2) at a 50 cm spatial resolution and 0.05°C thermal resolution. At present, the only way to generate a large area, high-spatial resolution TIR scene is to acquire separate airborne flight lines and mosaic them together. However, the ambient sensed temperature within, and between flight lines naturally changes during acquisition (due to varying atmospheric and local micro-climate conditions), resulting in mosaicked images with different temperatures for the same scene components (e.g. roads, buildings), and mosaic join-lines arbitrarily bisect many thousands of homes. In combination these effects result in reduced utility and classification accuracy including, poorly defined HEAT Metrics, inaccurate hotspot detection and raw imagery that are difficult to interpret.
In an effort to minimize these effects, three new semi/automated post-processing algorithms (the protocol) are described, which are then used to generate a 43 flight line mosaic of TABI-1800 data from which accurate Calgary waste heat maps and HEAT metrics can be generated. These algorithms (presented as four peer-reviewed papers) - are: (a) Thermal Urban Road Normalization (TURN) - used to mitigate the microclimatic variability within a thermal flight line based on varying road temperatures; (b) Automated Polynomial Relative Radiometric Normalization (RRN) - which mitigates the between flight line radiometric variability; and (c) Object Based Mosaicking (OBM) - which minimizes the geometric distortion along the mosaic edge between each flight line. A modified Emissivity Modulation technique is also described to correct H-res TIR images for emissivity. This combined radiometric and geometric post-processing protocol (i) increases the visual agreement between TABI-1800 flight lines, (ii) improves radiometric agreement within/between flight lines, (iii) produces a visually seamless mosaic, (iv) improves hot-spot detection and landcover classification accuracy, and (v) provides accurate data for thermal-based HEAT energy models.
Description
Keywords
Geography
Citation
Rahman, M. M. (2014). Developing a semi/automated protocol to post-process large volume, High-resolution airborne thermal infrared (TIR) imagery for urban waste heat mapping (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/25617