Image segmentation method of slime floatation foam image based on machine vision
XIE Peihong
XU Shoukun
LU Zhaolin
ZHOU Changchun
Abstract：An image segmentation method based on ROF model and top-hat transformation of mathematical morphology was proposed to solve the problems of noise interference and adherent foam in slime flotation foam image.The discontinuous function included in the func-tion space of bounded total variation of the ROF model corresponded to the edge profile of the objective in the image.Hence,the ROF mod-el was ideally suited for the image filtering.The space among foams could be increased using the top-hat transformation.Adherent foams could be segmented easily.The image segmentation method was conducted as follows.First,the slime flotation foam image was filtered u-sing the ROF model.Then,the image was top-hat transformed and the space increased accordingly.Finally,the watershed algorithm was applied to segment the foam image.Theoretical analysis and experimental results showed that the method could segment the foam fast and accurately in complex slime flotation foam image,the accuracy could reach 81.1%.
Keywords：Rudin-Osher-Fatemi modeltop-hat transformationslime flotationfoam imageimage segmentation
Publication Date：2015-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：4( 42-45 )