INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
CLASSIFICATION OF ULTRASOUND IMAGES FOR THYROID DETECTION USING SVM
Authors Name:
N VIGNESH
, A.MOHAMED ASHFAQ , D DHARANIBALAN , A BHUVANESHWARI , R RADHIGA
Unique Id:
IJSDR2105012
Published In:
Volume 6 Issue 5, May-2021
Abstract:
Nowadays the health issue is being worried a lot and so the workload of a doctor becomes comparatively huge. In current scenario, unidentified thyroid has lots of serious medical issues. So, to reduce the workload, the image segmentation and classification of thyroid ultrasound images is most necessary. In our proposed project, the thyroid ultrasound images are taken for further processing. The US images first undergo a pre-processing stage, involving processes like grayscale conversion, intensity calculation and histogram equalization. Most commonly, GLCM algorithm is being used for the segmentation of ultrasound images. The feature output that is obtained is then applied to the PCA as input. All the features extracted from each data are combined into a single matrix for classification with the help of PCA. For classification purposes, The classifier is SVM. In our proposal, classification is done by CNN because SVM is a binary classifier whereas CNN classifier is more efficient compared to SVM.
Keywords:
Thyroid Ulrasound Images, PCA, GLCM, SVM.
Cite Article:
"CLASSIFICATION OF ULTRASOUND IMAGES FOR THYROID DETECTION USING SVM", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.6, Issue 5, page no.69 - 72, May-2021, Available :http://www.ijsdr.org/papers/IJSDR2105012.pdf
Downloads:
000337348
Publication Details:
Published Paper ID: IJSDR2105012
Registration ID:192962
Published In: Volume 6 Issue 5, May-2021
DOI (Digital Object Identifier):
Page No: 69 - 72
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn