Cite this DOI

Prakash, B. (2019). Algorithm for Skin Lesion Segmentation. *Journal of Science & Technology*, *04*(02), 20–29. https://doi.org/10.46243/jstj.2019.v4.i2.122

@article{prakash2019algorithm,
  author    = {Prakash, B.L},
  title     = {{Algorithm for Skin Lesion Segmentation}},
  journal   = {Journal of Science \& Technology},
  year      = {2019},
  month     = {mar},
  volume    = {04},
  number    = {02},
  pages     = {20--29},
  publisher = {Longman Publishers},
  issn      = {2456-5660},
  doi       = {10.46243/jstj.2019.v4.i2.122},
  url       = {https://doi.org/10.46243/jstj.2019.v4.i2.122},
  language  = {en},
  abstract  = {In most recent applications of image analysis, the difficulty faced is to detect proper structure of an irregularly shaped object. This is mostly seen in the applications of medical field such as skin lesion segmentation. It is also a critical task to determine the exact border line of the lesion. Also early detection of skin cancer is an essential problem in the recent years of development in image processing. There are many types of skin lesion appearances. Some of them are blurred, some are irregular in shape, some are on dark skin, and some are seen with lots of hair on the skin. The main aim of this research paper is to detect the malignant region of skin lesion, identify its stage, and segment it out from the skin image. During the process, it is essential to preprocess the input image by performing conversion of a color image to gray scale image, removal of blur, removal of noise, smoothing of images, etc. It also involves grouping the similar pixels into one cluster, likewise obtaining various clusters based on similarities. Then it is essential to perform the extraction of features of the lesion and displaying the segmented lesion. In the current research a novel approach is developed to obtain the clusters of any shape and is tested on skin lesion images for detecting the cancer cells. Performance of developed clustering algorithm is tested by measuring various parameters based on distance metric and few similarity indexes. The proposed method is also compared to other approaches that are developedearlier.}
}

TY  - JOUR
TI  - Algorithm for Skin Lesion Segmentation
AU  - Prakash, B.L
JO  - Journal of Science & Technology
PY  - 2019
DA  - 2019/03/01/
VL  - 04
IS  - 02
SP  - 20
EP  - 29
PB  - Longman Publishers
SN  - 2456-5660
LA  - en
AB  - In most recent applications of image analysis, the difficulty faced is to detect proper structure of an irregularly shaped object. This is mostly seen in the applications of medical field such as skin lesion segmentation. It is also a critical task to determine the exact border line of the lesion. Also early detection of skin cancer is an essential problem in the recent years of development in image processing. There are many types of skin lesion appearances. Some of them are blurred, some are irregular in shape, some are on dark skin, and some are seen with lots of hair on the skin. The main aim of this research paper is to detect the malignant region of skin lesion, identify its stage, and segment it out from the skin image. During the process, it is essential to preprocess the input image by performing conversion of a color image to gray scale image, removal of blur, removal of noise, smoothing of images, etc. It also involves grouping the similar pixels into one cluster, likewise obtaining various clusters based on similarities. Then it is essential to perform the extraction of features of the lesion and displaying the segmented lesion. In the current research a novel approach is developed to obtain the clusters of any shape and is tested on skin lesion images for detecting the cancer cells. Performance of developed clustering algorithm is tested by measuring various parameters based on distance metric and few similarity indexes. The proposed method is also compared to other approaches that are developedearlier.
DO  - 10.46243/jstj.2019.v4.i2.122
UR  - https://doi.org/10.46243/jstj.2019.v4.i2.122
ER  -

{
    "type": "article-journal",
    "id": "10.46243/jstj.2019.v4.i2.122",
    "DOI": "10.46243/jstj.2019.v4.i2.122",
    "URL": "https://doi.org/10.46243/jstj.2019.v4.i2.122",
    "title": "Algorithm for Skin Lesion Segmentation",
    "source": "Smart Scholars DOI Registry",
    "container-title": "Journal of Science & Technology",
    "author": [
        {
            "family": "Prakash",
            "given": "B.L"
        }
    ],
    "issued": {
        "date-parts": [
            [
                2019,
                3,
                1
            ]
        ]
    },
    "volume": "04",
    "issue": "02",
    "page": "20-29",
    "publisher": "Longman Publishers",
    "language": "en",
    "abstract": "In most recent applications of image analysis, the difficulty faced is to detect proper structure of an irregularly shaped object. This is mostly seen in the applications of medical field such as skin lesion segmentation. It is also a critical task to determine the exact border line of the lesion. Also early detection of skin cancer is an essential problem in the recent years of development in image processing. There are many types of skin lesion appearances. Some of them are blurred, some are irregular in shape, some are on dark skin, and some are seen with lots of hair on the skin. The main aim of this research paper is to detect the malignant region of skin lesion, identify its stage, and segment it out from the skin image. During the process, it is essential to preprocess the input image by performing conversion of a color image to gray scale image, removal of blur, removal of noise, smoothing of images, etc. It also involves grouping the similar pixels into one cluster, likewise obtaining various clusters based on similarities. Then it is essential to perform the extraction of features of the lesion and displaying the segmented lesion. In the current research a novel approach is developed to obtain the clusters of any shape and is tested on skin lesion images for detecting the cancer cells. Performance of developed clustering algorithm is tested by measuring various parameters based on distance metric and few similarity indexes. The proposed method is also compared to other approaches that are developedearlier.",
    "ISSN": "2456-5660"
}

From the record as registered (version 2) — the record and its history. Programs: https://registry.smartscholars.in/api.php?action=cite&doi=10.46243%2Fjstj.2019.v4.i2.122 gives all four in one JSON answer.