A Semantic Approach to Segmentation of Overlapping Objects

Summary

Objectives: This paper aims at introducing a novel approach for segmentation of overlapping objects and at demonstrating its applicability to medical images.

Methods: This work details a novel approach enhancing the known theory of full-segmentation of an image into regions by lifting it to a semantic segmentation into objects. Our theory allows the formal description of partitioning an image into regions on the first level and allowing the occurrence of overlaps and occlusions of objects on a second, semantic level. Possible applications for the use of this ‘semantical segmentation‘ are the analysis of radiographs and micrographs. We demonstrate our approach by the example of segmentation and separation of overlapping cervical cells and cell clusters on a set of 787 image pairs of registered PAP- and DAPI-stained micrographs. The semantical cell segmentation yielding areas of cell plasmas and nuclei are compared to a manual segmentation of the same images, where 2212 cells have been labeled. A direct comparison of over and under-segmentation between the two segmentation sets yields a mean difference value of 10.15% for the nuclei and 10.80% for the plasma.

Results: Using the proposed theory of semantical segmentation of images in combination with adequate models of the image contents, our approach allows identifying, separating and distinguishing several overlapping, occluding objects in medical images. Applying the proposed theory to the application of cervical cell segmentation from overlapping cell clusters and aggregates, it can be seen that it is possible to formally describe the complex image contents.

Conclusions: The proposed method of semantical segmentation is a mighty tool and under the assumption of the subtractive transparency model can be used in different medical image processing applications such as radiology and microscopy. By using alternative models to solve the ambiguities attached to overlaps and occlusions, further fields of application can be addressed.

Keywords

Algorithms - computer-assisted image processing - computer-assisted image interpretation - photomicrography - fluorescence microscopy - image segmentation - overlapping objects - occluding objects

• References

• 1 Li SZ. Recognizing multiple overlapping objects in image: An optimal formulation. IEEE Trans Image Proc 2000; 9 (02) 273-6.
• 2 Gonzales RC, Woods RE. Digital Image Processing. Reading, MA, USA: Addison Wesley; 1993
• 3 Paulus D. Aktives Sehen. Osnabrück: Der andere Verlag. 2001
• 4 Anderson BL, Julesz B. A theoretical analysis of illusory contour formation in stereopsis. Psychological Review 1995; (195) 705-43.
• 5 Fernandez G, Zyrd JP, Kunt M. A New Plant Cell Image Segmentation Algorithm. In Braccini C, DeFloriani L, Vernazza G. (eds) Image Analysis and Processing, 8th International Conference, ICIAP ’95, San Remo, Italy, September 13-15, 1995, Proceedings, LNCS vol. 974. 1995. Berlin: Springer; 229-34.
• 6 Jarkrans T. Algorithms for Cell Image Analysis in Cytology and Pathology. PhD-thesis, Faculty of Science and Technology, Uppsala University, Uppsala, Sweden. 1996
• 7 Byriel J. Neuro-Fuzzy Classification of Cells in Cervical Smears. Masters sc. thesis, Technical University of Denmark, Lyngby, Feb. 1999
• 8 Anoraganingrum D, Kröner S, Gottfried B. Cell segmentation with adaptive region growing. In: Int. Conf. on Image Analysis and Processing, Venice, Italy, Sep. 27-29. 1999
• 9 Anoraganingrum D. Cell image segmentation: Global vs. local adaptive segmentation. In: Proc ISSM Kassel, Germany, Oct. 8-9. 1999
• 10 Wu K, Gauthier D, Levine M. Live cell image segmentation. Technical report, Center for Intelligent Machines, McGill University, Montreal, Quebec, Canada, Oct. 1995
• 11 Bamford P, Lovell B. A Water Immersion Algorithm for Cytological Image Segmentation. In Proc of Segment ’96, Sydney, Dec. 1996
• 12 Wählby C, Lindblad J, Vondrus M, Bengtsson E, Björksten L. Algorithm for cytoplasm segmentation of fluorescence labelled cells. Anal Cell Pathol 2002; 24: 101-11.
• 13 Lezoray O, Cardot H. Cooperation of color pixel classification schemes and color watershed: a study for microscopical images. IEEE Trans Image Proc 2002; 11 (07) 783-9.
• 14 Malpica N, de Solorzano CO, Vaquero JJ, Vallcorba I, Garcia-Sagredo JM, del Pozo F. Applying Watershed Algorithms to the Segmentation of Clustered Nuclei. Cytometry 1997; 28: 289-97.
• 15 Bamford P, Lovell B. Improving the robustness of cell nucleus segmentation. In: Proc of the 9th British Machine Vision Conf., Southampton, UK. 1998
• 16 Meas-Yedid V, Cloppet F, Roumier A, Alcover A, Olivo-Marin JC, Stamon G. Quantitative microscopic image analysis by active contours. In: Proc IAPR-Vision Interface, Ottawa, Canada. 2001: 277-84.
• 17 Mackin R, Roysam B, Turner J. Adaptive 3-d segmentation algorithms for microscope images using local infocus, and contrast features: application to pap smear. In: IEEE Proc International Conference on Image Processing, Washington D.C. (USA), Oct 23-26 1995; vol 3: 160-3.
• 18 Soille P. Morphological Image Analysis: Principles and Applications. Berlin: Springer-Verlag; 1999
• 19 Metzler V, Bienert G, Lehmann T, Mottaghy M, Spitzer K. A novel method for quantifying shape deformations applied to biocompatibility testing. ASAIO Journal 1999; 45 (04) 264-71.
• 20 Metzler V, Lehmann T, Bienert H, Mottaghy K, Spitzer K. Scale-independent shape analysis for quantitative cytology using mathematical morphology. Computers in Medicine and Biology 2000; 30: 135-51.
• 21 Papanicolaou G. A new procedure for staining vaginal smears. Science 1942; 95: 438-9.
• 22 Grobe M, Volk H, Münzenmayer C, Wittenberg T. Segmentierung von überlappenden Zellen in Fluoreszenz- und Durchlichtaufnahmen. In Bildverarbeitung für die Medizin 2003. Heidelberg: Springer; 2003: 201-5.
• 23 Paulus D, Hornergger J. Pattern Recognition of Images and Speech in C++. Wiesbaden: Vieweg; 1997
• 24 Nauth HF. Gynäkologische Zytodiagostik. Stuttgart: Thieme Verlag; 1997
• 25 Freudenberg N, Ross A, Kortsek C. Grundlagen der Zytopatologie: Kurzlehrbuch und Atlas der allgemeinen und speziellen Zytodiagnostik. Freiburg: Karger; 2002
• 26 Otsu N. A threshold selection method from gray-level histograms. IEEE Transactions on Systems, Man, and Cybernetics 1979; 9 (01) 62-6.
• 27 Hoover A, Jean-Baptiste G, Jiang X, Flynn P, Bunke H, Goldgof D, Bowyer K, Eggert D, Fitzgibbon A, Fisher R. An experimental comparison of range image segmentation algorithms. IEEE Pami 1996; 18 (07) 673-89.
• 28 Martin F, Fowlkes C, Tal D, Malik J. A Database of Human Segmented Natural Images and its Application to Evaluating Sementation Algorithms and Measuring Ecological Statistics. In: Proc ICCV, Vancouver, July. 2001: 416-25.