Bayesian inference and sidescan restoration
Abstract
We consider a Bayesian approach to the problem of inferring parameters of the SONAR environment given only the gathered sidescan image. A simplified model of the process is developed along with suitable distributions on the parameters, and a sampling technique is utilised to estimate the parameters most likely given the data. As an example, we apply this technique to estimation of a step-gain TVG curve and use the results in restoration of legacy sidescan data.
Report Number
CP-45Source
In: High Frequency Seafloor Acoustics (SACLANTCEN Conference Proceedings CP-45), 1997, pp. 91-98Date
1997Author(s)
Calder, B. R.
; Linnett, L. M.