A Regions of Confidence Based Approach to Enhance Segmentation with Shape PriorsReportar como inadecuado

A Regions of Confidence Based Approach to Enhance Segmentation with Shape Priors

A Regions of Confidence Based Approach to Enhance Segmentation with Shape Priors - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

We propose an improved region based segmentation model with shape priors that uses labels of confidence-interest toexclude the influence of certain regions in the image that may not provide useful information for segmentation. Thesecould be regions in the image which are expected to have weak, missing or corrupt edges or they could be regions in theimage which the user is not interested in segmenting, but are part of the object being segmented. In the training datasets,along with the manual segmentations we also generate an auxiliary map indicating these regions of low confidence-interest.Since, all the training images are acquired under similar conditions, we can train our algorithm to estimate these regions aswell. Based on this training we will generate a map which indicates the regions in the image that are likely to contain nouseful information for segmentation. We then use a parametric model to represent the segmenting curve as a combinationof shape priors obtained by representing the training data as a collection of signed distance functions. We evolve anobjective energy functional to evolve the global parameters that are used to represent the curve. We vary the influenceeach pixel has on the evolution of these parameters based on the confidence-interest label. When we use these labels toindicate the regions with low confidence; the regions containing accurate edges will have a dominant role in the evolutionof the curve and the segmentation in the low confidence regions will be approximated based on the training data. Since ourmodel evolves global parameters, it improves the segmentation even in the regions with accurate edges. This is because weeliminate the influence of the low confidence regions which may mislead the final segmentation. Similarly when we usethe labels to indicate the regions which are not of importance, we will get a better segmentation of the object in the regionswe are interested in.

Laboratory of Computational Computer Vision Publications -

Autor: Appia, Vikram V. - Ganapathy, Balaji - Abufadel, Amer - Yezzi, Anthony - Faber, Tracy - -

Fuente: https://smartech.gatech.edu/

Documentos relacionados