Abstract—This paper extends the classical segmentation method known as region growing using a new localization method which uses algorithms from computer science and computer vision. Described method was successfully applied to the problem of cell localization and segmentation. The method proves to be useful as part of a whole autonomous segmentation process. Furthermore, methods from spatial statistics were employed in order to model the spatial distribution of the localized objects. In-depth discussion of these methods including density plots and the Ripley’s K function is presented along with the results for test set used in the experiments.
Index Terms—Object localization, segmentation, spatial analysis
The authors are with the University of Essex, Colchester, UK (e-mail: gebost@essex.ac.uk, bbosta@essex.ac.uk).
[PDF]
Cite:Erkan Bostanci and Betul Bostanci, "Object Localization Based Segmentation and Spatial Analysis Using Computer Vision," International Journal of Computer Theory and Engineering vol. 5, no. 3, pp. 488-493, 2013.
