Bayesian Image Classiication with Baddeley's Delta Loss
نویسنده
چکیده
In this paper we adopt Baddeley's delta metric (Baddeley 1992) as a loss function in Bayesian image restoration and classiication. We develop a new algorithm that allows to estimate the corresponding optimal Bayesian estimator, for which good practical estimates can be obtained by approximately the same cost as traditional estimators like the Marginal Posterior Mode (MPM). A comparison of such classiication with MPM shows signiicant advantages, especially with respect to ne structures.
منابع مشابه
Bayesian Inference and Optimization Strategies for Some Detection and Classification Problems in Sonar Imagery
In this paper, we investigate the use of the Bayesian inference for some detection and classiication problems of great importance in sonar imagery. More precisely this paper is concerned with the segmentation of sonar image, the classiication of object lying on the sea-bottom and the classiication of sea-oor. These aforementioned classiication tasks are based on the identiication of the detecte...
متن کاملPerformance Measures for Nn-classiiers
This paper introduces Bayesian classiication, measurement of classiication costs by loss functionals and methods for estimation of loss when the true classiication rules are unknown. Cross-validation and several bootstrap methods are reviewed and compared using two artiicial examples.
متن کاملRobust Partial Volume Tissue Classiication of Cerebral Mri Scans
In Magnetic Resonance Images (MRI) a voxel may contain multiple tissue types (partial volume eeect). We concentrate in the classiication of these voxels using an adaptive Bayesian approach and pay particular attention to practical implementation problems induced by the modeling of partial volume voxels. Moreover, we show that this algorithm is suitable to perform tissue classiication of brain M...
متن کاملImproving the Performance of Boosting for Naive Bayesian Classification
This paper investigates boosting naive Bayesian classiica-tion. It rst shows that boosting cannot improve the accuracy of the naive Bayesian classiier on average in a set of natural domains. By analyzing the reasons of boosting's failures, we propose to introduce tree structures into naive Bayesian classiication to improve the performance of boosting when working with naive Bayesian classiicati...
متن کاملImproving the Performance of Boosting forNaive Bayesian Classi cationKai
This paper investigates boosting naive Bayesian classiica-tion. It rst shows that boosting cannot improve the accuracy of the naive Bayesian classiier on average in a set of natural domains. By analyzing the reasons of boosting's failures, we propose to introduce tree structures into naive Bayesian classiication to improve the performance of boosting when working with naive Bayesian classiicati...
متن کامل