Touching particles separation based on area ratio of circular mask

In this paper a method for separation of partially overlapping particle is proposed based on circular mask and the framework including image segmentation concave points detection point pair matching and shape estimation is realized. First the color image is preprocessed to obtain a binary image. Then the comer detection algorithm is adopted to achieve rough estimation on potential concave points and the area ratio method was used to select the candidate points so as to obtain the segmentation points. Since then the obtained segmentation points are matched to each pairs by gray projection according to a certain criteria. Finally each pair of points and one of their adjacent pixels on the border is used to estimate the status of individual particles before adhesion and the contour of each particle is outputted. Experiments show that the proposed method is consistent with the human visual perception and has achieved good results.

[1]  Andrew G. Dempster,et al.  Analysis of infected blood cell images using morphological operators , 2002, Image Vis. Comput..

[2]  Sim Heng Ong,et al.  A rule-based approach for robust clump splitting , 2006, Pattern Recognit..

[3]  Mohamed Cheriet,et al.  Automatic segmentation of cells from microscopic imagery using ellipse detection , 2007 .

[4]  Fabio A. González,et al.  Automatic Clump Splitting for Cell Quantification in Microscopical Images , 2007, CIARP.

[5]  Chen Hao.  Research of automatically separating algorithm for overlap cell based on searching concave spot , 2007 .

[6]  Changming Sun,et al.  Splitting touching cells based on concave points and ellipse fitting , 2009, Pattern Recognit..

[7]  Jagath C. Rajapakse,et al.  Segmentation of Clustered Nuclei With Shape Markers and Marking Function , 2009, IEEE Transactions on Biomedical Engineering.

[8]  Jiandeng Huang.  An improved algorithm of overlapping cell division , 2010, 2010 International Conference on Intelligent Computing and Integrated Systems.

[9]  Chanho Jung,et al.  Segmenting Clustered Nuclei Using H-minima Transform-Based Marker Extraction and Contour Parameterization , 2010, IEEE Transactions on Biomedical Engineering.

[10]  Christophoros Nikou,et al.  Overlapping Cell Nuclei Segmentation Using a Spatially Adaptive Active Physical Model , 2012, IEEE Transactions on Image Processing.

[11]  Javier DeFelipe,et al.  Segmentation of neuronal nuclei based on clump splitting and a two-step binarization of images , 2013, Expert Syst. Appl..

[12]  Yu Ding,et al.  Segmentation, Inference and Classification of Partially Overlapping Nanoparticles , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Olli Yli-Harja,et al.  A novel method for splitting clumps of convex objects incorporating image intensity and using rectangular window-based concavity point-pair search , 2013, Pattern Recognit..

[14]  Osman Kalender,et al.  Automatic segmentation, counting, size determination and classification of white blood cells , 2014 .

[15]  Narendra Ahuja,et al.  Automatic segmentation of granular objects in images: Combining local density clustering and gradient-barrier watershed , 2014, Pattern Recognit..

[16]  Gustavo Carneiro,et al.  An Improved Joint Optimization of Multiple Level Set Functions for the Segmentation of Overlapping Cervical Cells , 2015, IEEE Transactions on Image Processing.