PROJECT TITLE :
Non-Gaussian Target Detection in Sonar Imagery Using the Multivariate Laplace Distribution
This paper introduces a replacement non-Gaussian detection methodology for advanced-valued artificial aperture sonar (SAS) imagery. The detection methodology is based on a multivariate extension of the Laplace distribution derived employing a scale mixture of Gaussian distributions. A goodness-of-fit check in the form of a chance ratio is then conducted on a sonar imagery knowledge set consisting of high-frequency (HF) and broadband (BB) images coregistered over the same region on the seafloor showing the proposed model's applicability in sonar imagery. Detection based mostly on testing the equality of parameters from two populations is then implemented on a database containing actual SAS pictures of the seafloor with synthetically generated targets inserted into the pictures and compared to an analogous non-Gaussian technique. Detection performance in this paper is given in terms of receiver-operator characteristic (ROC) curve attributes, chance of detection, and average false alarm rate.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here