Evaluation of Algorithms for Sequential Decision in Biosonar Target Classification

A sequential decision problem, based on the task of identifying the species of trees given acoustic echo data collected from them, is considered with well-known stochastic classifiers, including single and mixture Gaussian models. Echoes are processed with a preprocessing stage based on a model of mammalian cochlear filtering, using a new discrete low-pass filter characteristic. Stopping time performance of the sequential decision process is evaluated and compared. It is observed that the new low pass filter processing results in faster sequential decisions. Keywords-- Classification, neuro-spike coding, parametric model, Gaussian mixture with EM algorithm, sequential decision.

• Type: Conference
• Year: 2004
• Where: IJIT
• Authors: Turgay Temel, John Hallam
• Keywords: Artificial Intelligence, IJIT 2004, Sequential Decision, Sequential Decision Problem, Sequential Decision Process