Effects of atmospheric correction on vehicle classification with single and dual band infrared images

Download
2014
Özsaraç, Seçkin
A vehicle classification system, which uses features based on radiometry, is developed for both single band and dual band infrared (IR) image sequences. For classification using dual band sequences, the process is divided into six components. These are registration, fusion, moving vehicle detection, geometry estimation, atmospheric effects removal, and classification. In the single band case, registration and fusion steps are not used. The first major novelty of the thesis is an atmospheric correction, i.e. atmospheric effects removal, system that considers the spectral characteristics of the detector, lens, and filter. In this system, an enhanced temperature calibration method is developed and it is shown that the temperature accuracy for the dynamic range of the IR camera is very close to the ultimate goal, i.e. Noise Equivalent Temperature Difference (NETD) value of the camera. Furthermore, as the atmospheric effects vary from pixel to pixel, a geometry estimation method is developed to estimate the Line Of Sight (LOS) geometry for each pixel using only the Global Positioning System (GPS) coordinates of the camera and a Point Of Interest (POI) in the scene. The second major novelty of the thesis lies in the usage of the atmospherically corrected radiance values as features to improve the classification performance of the detected objects. The motivation is, each vehicle class has a discriminating radiance value that originates from the source temperature of the object modified by the intrinsic characteristics of the radiating surface. As a consequence, significant performance gains are achieved due to the use of radiance values in classification both for a single band and a dual band measurement systems.
Citation Formats
S. Özsaraç, “Effects of atmospheric correction on vehicle classification with single and dual band infrared images,” Ph.D. - Doctoral Program, Middle East Technical University, 2014.