Abstract
Skin cancer is considered to be the deadliest disease. Lesion is a suspicious part which has an unusual growth compared to skin and also appears as a smooth surface with size variation, indiscriminate shape, and unusual colors. Segmentation plays an essential and crucial role here. When an image is divided into segments, important features can be projected and processed instead of complete image. Some expert dermatologists can see the segmented part of lesion and conclude the chances of occurring melanoma and non-melanoma. This phase plays a crucial role in early diagnosis and detection of cancer. However, selecting an apt segmentation technique for various data set images is a major challenge in the medical field. Hence, this work addresses a selection of suitable segmentation method which has to confer a good result. In this paper, three approaches of segmentation techniques binary Otsu, marker-based watershed, and K-means clustering are implemented and compared especially for irregular border lesion. Segmentation results are evaluated based on quality assessment metrics of image such as mean square error, mean absolute error, structural similarity index, and peak signal-to-noise ratio, i.e., MSE, MAE, SSIM, and PSNR. On an average, it is observed that for marker-based watershed segmentation MSE and MAE values are reduced to 35% and 10%. PSNR values are increased to 15%, and SSIM has shown an increase of 70% when compared to other two methods. This research shows that marker-based watershed segmentation works well for irregular border lesion images of melanoma.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Gutman D, Codella NC, Celebi E, Helba B, Marchetti M, Mishra N, Halpern A (2016) Skin lesion analysis toward melanoma detection: a challenge at the international symposium on biomedical imaging (ISBI) 2016, hosted by the international skin imaging collaboration (ISIC). arXiv preprint arXiv:1605.01397
Jacily Jemila S, Brintha Therese A (2019) Selection of suitable segmentation technique based on image quality metrics. Imaging Sci J 67(8):475–480
Ali AR, Li J, Yang G (2020) Automating the ABCD rule for melanoma detection: a Survey. IEEE Access 8:83333–83346. https://doi.org/10.1109/ACCESS.2020.2991034
Chakkaravarthy Prabhu A, Chandrasekar A (2019) Automatic detection and segmentation of melanoma using fuzzy c-means. In: 2019 Fifth international conference on science technology engineering and mathematics (ICONSTEM). IEEE, New York, pp 132–136
Manikandan LC, Selvakumar RK, Nair S, Anu H, Sanal Kumar KP (2021) Hardware implementation of fast bilateral filter and canny edge detector using Raspberry Pi for telemedicine applications. J Amb Intell Hum Comput 12(5):4689–4695
Zaini SZS, Marzuki NNSM, Abdullah MF, Ahmad KA, Isa Sulaiman SN (2019) Image quality assessment for image segmentation algorithms: qualitative and quantitative analyses. In: 9th IEEE International conference on control system, computing and engineering (ICCSCE). IEEE, Penang, Malaysia, pp 66–71
Sreedhar B, Manjunath Swamy BE, Sunil Kumar M (2020) A comparative study of melanoma skin cancer detection in traditional and current image processing techniques. In: Proceedings of the fourth international conference on I-SMAC (IoT in social, mobile, analytics and cloud). IEEE, New York, pp 654–658
Broti T, Siddika A, Rituparna S, Hossain N, Sakib N (2020) Medical image analysis system for segmenting skin diseases using digital image processing technology. Int J Appl Inf Syst 12(28):7–15
Jamil U, Sajid A, Hussain M, Aldabbas O, Shafiq Afshan Alam Umair M (2019) Melanoma segmentation using bio-medical image analysis for smarter mobile healthcare. Springer 10(10):4099–4120
Yuan C, Yang H (2019) Research on K-value selection method of K-means clustering algorithm. Multidisciplinary Sci J 2(2):226–235
Kaur R, Maini R (2020) Evaluation and analysis of edge detection techniques on Leukemia images. Adv Math: Sci J 9(6):3721–3732
Metib MH, Abdulhssien MF, Abdulmunem AA (2020) Skin dermatitis detection using image segmentation techniques. In: 2nd International scientific conference of Al-Ayen University (ISCAU-2020), IOP conference series: materials science and engineering, vol 928. IOP Publishing, pp 1–9. https://doi.org/10.1088/1757-899X/928/3/032018
Sara U, Akter M, Uddin MS (2019) Image quality assessment through FSIM, SSIM, MSE and PSNR-a comparative study. J Comput Commun 7(3):8–18
Gothi S, Baraskar R, Agrawal S (2019) An efficient approach of image segmentation for skin cancer detection. Int J Sci Technol Res 7(2):783–787
Shanthi V, Sridevi G, Charanya R, Josphin Mary R (2020) Watershed algorithm in multichannel for skin lesion segmentation. Euro J Mol Clin Med 7(9):1374–1378
Zaw MT (2018) Than HTIKE AUNG: automatic segmentation of skin lesion in dermoscopic images. Int J Sci Eng Technol Res 8(9):0223–0229
Mwawado RH, Maiseli BJ, Dida M (2020) Robust edge detection method for the segmentation of diabetic foot ulcer images. Eng Technol Appl Sci Res 10(4):6034–6040
ISIC Homepage. https://www.isic-archive.com
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Gnana Mayuri, K., Sathish Kumar, L. (2023). Assessment of Segmentation Techniques for Irregular Border Lesion Images in Melanoma. In: Buyya, R., Hernandez, S.M., Kovvur, R.M.R., Sarma, T.H. (eds) Computational Intelligence and Data Analytics. Lecture Notes on Data Engineering and Communications Technologies, vol 142. Springer, Singapore. https://doi.org/10.1007/978-981-19-3391-2_12
Download citation
DOI: https://doi.org/10.1007/978-981-19-3391-2_12
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-3390-5
Online ISBN: 978-981-19-3391-2
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)