Welcome to the COVID-19 Detection Using Chest X-rays project! This repository contains a comprehensive solution for detecting COVID-19 from chest X-ray images using advanced image processing and machine learning techniques. Our approach leverages state-of-the-art methods to enhance, preprocess, and extract features from X-ray images, followed by the application of various machine learning models to accurately classify the images as COVID-positive or COVID-negative.
- Image Visualization: Display and compare COVID and Non-COVID X-ray images.
- Image Enhancement: Improve image quality using histogram equalization.
- Data Preprocessing: Normalize and augment images for better model performance.
- Feature Extraction: Extract meaningful features using Histogram of Oriented Gradients (HOG).
- Model Training: Train multiple machine learning models including SVM, Logistic Regression, Random Forest, and more.
- Model Evaluation: Assess the performance of models using accuracy, confusion matrix, and classification reports.
- Model Persistence: Save and load trained models for future use.
Ensure you have the following installed:
- Python 3.x
- Required Python libraries (listed in
requirements.txt)
-
Clone the repository:
git clone https://github.com/your-username/covid-xray-detection.git cd covid-xray-detection -
pip install -r requirements.txt
you can download the dataset from here:-https://data.mendeley.com/datasets/xztwjmktrg/2
The following table shows the accuracy of different models:
| Model | Accuracy |
|---|---|
| Logistic Regression | 95% |
| Support Vector Machine | 97% |
| Random Forest | 94% |
| K-Nearest Neighbors | 93% |
| Decision Tree | 92% |
| Naive Bayes | 90% |
- This project is inspired by the need for accurate and efficient COVID-19 detection.
- Special thanks to the open-source community for providing valuable tools and resources.
For any queries or suggestions, please reach out to b22ch021@iitj.ac.in.