We present a new technique for estimating the sea surface heat flux from infrared image sequences. Based on solving an extension to the standard brightness change constraint equation in a total least squares (TLS) sense, the total derivative of the sea surface temperature with respect to time is obtained. Due to inevitable reflexes in field data the TLS framework was further extended to a robust estimation based on a Least Median of Squares Orthogonal Distances (LMSOD) scheme. From this it is possible for the first time to compute accurate heat flux densities to a high temporal and spatial resolution. Results obtained at the Heidelberg Aeolotron showed excellent agreement to ground truth and field data was obtained on the GasExII experiment.
CITATION STYLE
Garbe, C. S., & Jähne, B. (2001). Reliable estimates of the sea surface heat flux from image sequences. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2191, pp. 194–201). Springer Verlag. https://doi.org/10.1007/3-540-45404-7_26
Mendeley helps you to discover research relevant for your work.