Havza Bölütlemesinin Biyomedikal Görüntüler Üzerinde Bir Eğitim Aracı Olarak Uygulanması

Öz

Görüntü bölütleme, bir görüntüyü birden fazla bölüme ayırma işlemidir. Havza bölütlemesi, temel ve hızlı özellikleri nedeniyle yaygın olarak kullanılan bölge tabanlı bir bölütleme yöntemidir. Morfolojik işlemlere dayanan bu bölütleme yöntemi, özellikle manyetik rezonans görüntüleme (MRI), bilgisayarlı tomografi (CT) ve histopatolojik görüntüler gibi tıbbi görüntü uygulamalarında birçok alanda kullanılmaktadır. Genel olarak bu tür görüntülerde bölütleme problemleri gürültü, homojen olmama ve dokunan nesnelerden kaynaklanmaktadır. Bu tür problemleri çözmek için havza bölütlemesi güçlü bir araçtır. Tüm bu avantajlarının yanı sıra, havza bölütlemesi herhangi bir ek işlem yapılmadan görüntüye uygulanırsa, genellikle aşırı bölütleme problemi verir. Bu problemin üstesinden gelmek için işaretleyici kontrollü havza bölütlemesi geliştirilmiştir. Bu makalede, işaretleyici kontrollü havza bölütlemesi ile ilgili her biri beş farklı makaleden modifiye edilmiş beş farklı algoritma, gri tonlamalı bir kemik plazmasitom görüntüsü üzerinde uygulanmış ve sonuçlar karşılaştırılmıştır. MATLAB App Designer kullanılarak bu algoritmalarla ilgili bir eğitim platformu sunulmuştur. App Designer kullanılarak tasarlanan grafiksel kullanıcı arayüzü (GUI), işaretleyici kontrollü havza segmentasyonunun farklı algoritmalarını karşılaştırmak için yararlı bir araçtır. Ayrıca görüntü işleme öğrenenler bunu kolaylıkla kullanabilir ve algoritmaların etkinliğini gözlemleyebilirler. Dolayısıyla, oluşturulan eğitim platformu akademik ve eğitsel özelliklere sahiptir. Bu uygulamanın avantajı sadece farklı algoritmaları karşılaştırmak değil, aynı zamanda farklı işaretleyici kontrollü havza bölütlemesi yöntemlerini öğrenmektir.

Anahtar Kelimeler

• Görüntü işleme
• matematiksel morfoloji
• segmentasyon
• havza dönüşümü
• MATLAB GUI

Application of Watershed Segmentation on Biomedical Images as an Educational Tool

Öz

Image segmentation is the operation of dividing an image into multiple segments. Watershed segmentation is a widely used region-based segmentation method because of basic and fast features. This segmentation method that is based on morphological operations is used for several areas especially in medical image applications such as magnetic resonance imaging (MRI), computed tomography (CT) and histopathological images. In general, the segmentation problems with these types of images are originated from noise, nonhomogeneity and touching objects. For solving such problems, watershed segmentation is a powerful tool. In addition to all these advantages, if watershed segmentation is applied to image without any additional process, generally it gives an oversegmentation problem. Marker controlled watershed segmentation is improved to overcome this problem. In this article, five different algorithms about marker-controlled watershed segmentation, which each one is modified from five different articles, are applied on a grayscale bone plasmacytoma image and results are compared. An educational platform related with these algorithms is presented by using MATLAB App Designer. Designed graphical user interface (GUI) using App Designer is helpful tool for comparing different algorithms of marker-controlled watershed segmentation. Moreover, image processing learners can use it easily and can observe effectiveness of algorithms. So, the created educational platform has academic and educational characteristics. The advantage of this application is not only comparing different algorithms but also learning different marker-controlled watershed segmentation methods.

Anahtar Kelimeler

• Image processing
• mathematical morphology
• segmentation
• watershed transform
• MATLAB GUI

Kaynakça

1. Yumuş M, Apaydın M, Değirmenci A, Kaplanoğlu H, Kesikburun S, Karal Ö. Deep Convolutional Neural Networks Using SegNet for Automatic Spinal Canal Segmentation in Axial MRI. In: 2023 Innovations in Intelligent Systems and Applications Conference (ASYU); 11-13 October 2023; Sivas, Türkiye: IEEE. pp. 1-6.
2. Kilic C, Degirmenci A, Karal O. Segmentation of the Area Between Anterior and Posterior Vertebral Elements in Axial MR Images Using U-Net. In: 2024 Innovations in Intelligent Systems and Applications Conference (ASYU); 16-18 October 2024; Ankara, Türkiye: IEEE. pp. 1-6.
3. Beucher S, Meyer F. The morphological approach to segmentation: the watershed transformation. In: Mathematical Morphology in Image Processing. 1st ed. Boca Raton, FLA, USA: CRC Press, 1993.
4. Vincent L, Soille P. Watersheds in digital spaces: an efficient algorithm based on immersion simulations. IEEE Trans Pattern Anal Mach Intell 1991; 13(06): 583-598.
5. Gozalez RC, Woods RE. Digital Image Processing, 3rd ed. NY, USA: Pearson, 2009.
6. Meyer F, Beucher S. Morphological segmentation. J Visual Commun Image Represent 1990; 1(1): 21-46.
7. Beucher S. Use of watersheds in contour detection. In: Proc. Int. Workshop on Image Processing; 17-21 September 1979; Rennes, France. pp. 17-21.
8. Bieniek A, Moga A. An efficient watershed algorithm based on connected components. Pattern Recognit 2000; 33(6): 907-916.

9. Parvati K, Prakasa Rao BS, Mariya Das M. Image Segmentation Using Gray‐Scale Morphology and Marker‐Controlled Watershed Transformation. Discrete Dyn Nat Soc 2008; 2008(1): 384346.
10. Kaur M, Jindal G. Medical image segmentation using marker controlled watershed transformation. IJCST 2011; 2(4): 548-551.
11. Reza AW, Eswaran C, Dimyati K. Diagnosis of diabetic retinopathy: automatic extraction of optic disc and exudates from retinal images using marker-controlled watershed transformation. J Med Syst 2011; 35(6): 1491-1501.
12. El Allaoui A. Medical image segmentation by marker-controlled watershed and mathematical morphology. IJMA 2012; 4(3): 1-9.
13. Kalapala M, Rao VS, Srinivas K. Robust Tree Crown Delineation using Novel Marker Controlled Watershed Segmentation Algorithm. Int J Eng Res Technol 2012; 1(7):1-11.
14. Kaleem M, Sanaullah M, Hussain MA, Jaffar MA, Choi TS. Segmentation of brain tumor tissue using marker controlled watershed transform method. In: International Multi Topic Conference; 28-30 March 2012; Berlin, Heidelberg: Springer. pp. 222-227.
15. Koyuncu CF, Arslan S, Durmaz I, Cetin-Atalay R, Gunduz-Demir C. Smart markers for watershed-based cell segmentation. PloS one 2012; 7(11): e48664.
16. Yahya AA, Tan J, Hu M. A novel model of image segmentation based on watershed algorithm. Adv Multimedia 2013; 2013(1): 120798.
17. Beare R, Chen J, Adamson CL, Silk T, Thompson DK, Yang JYM, Andersan VA, Seal ML, Wood AG. Brain Extraction Using the Watershed Transform From Markers. Front Neuroinf 2013; 7(32): 1-15.
18. Sivagami M, Revathi T. Marker Controlled Watershed Segmentation Using Bit-Plane Slicing. IJIPVS 2013; 1(3):179-183.
19. Napoleon D, Santhoshi R, Shameena A. Verdict of Objects in Medical Images Using Marker-Controlled Watershed Image Segmentation. Int J Comput Appl Technol 2013; 975: 8887.
20. Ravi S, Khan AM. Bio-medical Image Segmentation Using Marker Controlled Watershed Algorithm: A Case Study. IJRET 2014; 3: 26-30.
21. Bandara, R. Image segmentation using unsupervised watershed algorithm with an over-segmentation reduction technique. arXiv preprint arXiv:1810.03908 2018.
22. Lahitha K, Amrytha R, Stafford M, Shivakumar DrM. Implementation of Watershed Segmentation. International Journal of Advanced Research in Computer and Communication Engineering 2016; 5(12): 196-199.
23. Lu Y, Jiang Z, Zhou T, Fu S. An Improved Watershed Segmentation Algorithm of Medical Tumor Image. In IOP Conference Series: Materials Science and Engineering; December 2019; 677 (4): IOP Publishing. pp. 042028.
24. Sarma R, Gupta YK. A comparative study of new and existing segmentation techniques. In IOP Conference Series: Materials Science and Engineering; January 2021; 1022(1): IOP Publishing. pp. 012027.
25. Dai Y, Meng L, Wang S, Sun F. A marker-controlled watershed algorithm for the intelligent picking of long jujubes in trees. Forests 2022; 13(7): 1063.
26. Guo Q, Wang Y, Yang S, Xiang Z. A method of blasted rock image segmentation based on improved watershed algorithm. Sci Rep 2022; 12(1): 7143.
27. Ghaderi S, Ghaderi K, Ghaznavi H. Using marker-controlled watershed transform to detect baker's cyst in magnetic resonance imaging images: A pilot study. J MED SIGNALS SENS 2022; 12(1): 84-89.
28. Salman SD, Bahrani AA. Segmentation of tumor tissue in gray medical images using watershed transformation method. Int J Adv Comp Techn 2010; 2(4): 123-127.
29. Goshal D, Acharjya PP. MRI image segmentation using watershed transform. Int J of Emerging Technology and Advanced Engineering 2012; 2(4): 373-376.
30. Solomon C, Breckon T. Fundamentals of Digital Image Processing: A practical approach with examples in Matlab. 1st ed., NJ, USA: John Wiley & Sons, 2011.
31. Vincent L. Morphological grayscale reconstruction in image analysis: applications and efficient algorithms. IEEE Trans Image Process 1993; 2(2): 176-201.
32. Chanda B, Kundu MK, Padmaja YV. A multi-scale morphologic edge detector. Pattern Recognition 1998; 31(10): 1469-1478.
33. Abdul-Nasir AS, Mashor MY, Mohamed Z. Modified global and modified linear contrast stretching algorithms: new colour contrast enhancement techniques for microscopic analysis of malaria slide images. Comput Math Methods Med 2012; 2012(1): 637360.
34. Jacobs D. Image Gradients. www.cs.umd.edu/~djacobs/CMSC426/ImageGradients.pdf Published 2005. Accessed 05 June 2023.
35. Beucher S. The watershed transformation applied to image segmentation. Scanning microscopy 1992; 1992(6): 299-314.
36. Sun H, Yang J, Ren M. A fast watershed algorithm based on chain code and its application in image segmentation. Pattern Recognit Lett 2005; 26(9): 1266-1274.
37. De Bock J, Philips W. Line segment based watershed segmentation. In: International Conference on Computer Vision/Computer Graphics Collaboration Techniques and Applications; March 2007; Berlin, Heidelberg: Springer Berlin Heidelberg. pp. 579-586.
38. Ruparelia S. Implementation of Watershed Based Image Segmentation Algorithm in FPGA. 2012. Master's Thesis.
39. Maheshwari D, Shah AA, Shaikh MZ, Chowdhry BS, Memon SR. Extraction of Brain Tumour in MRI Images Using Marker Controlled Watershed Transform Technique in MATLAB. J Biomed Eng Med Imaging 2015; 2(4): 9-16.
40. www.webpathology.com/ Accessed 13 May 2023.
41. Jung C, Kim C. Segmenting clustered nuclei using H-minima transform-based marker extraction and contour parameterization. IEEE Trans Biomed Eng 2010; 57(10): 2600-2604.
42. Chowdary PRV, Babu MN, Subbareddy TV, Reddy BM, Elamaran V. Image processing algorithms for gesture recognition using MATLAB. In: Proceedings of 2014 IEEE International Conference on Advanced Communication, Control and Computing Technologies; 08-10May 2014; ICACCCT 2014: IEEE. pp. 1511-1514.
43. Çankaya İ. MATLAB App Designer ile GUI Tasarımı ve Uygulamaları. Ankara, Türkiye: Seçkin Yayıncılık, 2024.