Bayesian Neural Networks for Image Analysis

We demonstrate the advantages of using Bayesian neural networks for image analysis. The Bayesian approa h provides onsistent way to do inferen e by ombining the eviden e from data to prior knowledge from the problem. A pra ti al problem with neural networks is to sele t the orre t omplexity for the model, i.e., the right number of hidden units or orre t regularization parameters. The Bayesian approa h o ers e ient tools for avoiding over tting even with very omplex models, and fa ilitates estimation of the on den e intervals of the results. In this ontribution we review the Bayesian methods for neural networks and present omparison results from ase studies in pro ess tomography and image segmentation. In the rst ase, neural networks were used to solve the inverse problem in ele tri al impedan e tomography. The Bayesian networks provided onsistently better results than other methods. In the se ond ase, the goal was to lo ate trunks of trees in forest s enes. With Bayesian network it was possible to use large number of potentially useful features and prior for determining the relevan e of the features automati ally.