Identification of Covid-19 patient with HOG-SVM technique
Goni, Md Osman (2025)
Goni, Md Osman
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025051311185
https://urn.fi/URN:NBN:fi:amk-2025051311185
Tiivistelmä
The COVID-19 pandemic, globally driven by the rapid spread of different variants of coronavirus, posed a severe threat to public health. Due to different reasons, such as a shortage of radiologists and other manpower in the COVID-19 department, the limited number of test kit availability on hospitals led to an inability to detect COVID-19 early that posed a greater threat to human life. Therefore, this thesis presents an easy and unique method to classify the coronavirus affected patient very easily using the patient’s chest x-ray images. This method proposes the HOG-SVM algorithm to analyse the x-ray images collect from the online COVID positive patients X-ray images to identify the cases. The experiment was evaluated using python programming languages and the TensorFlow library for the backend procedure while splitting 80% images as training and the rest 20% images as test images where 98.5% accuracy was acquired.