This project provides an easy-to-use interface where users can upload an MRI or CT scan image (PNG, JPG, JPEG). A YOLOv11 model, trained specifically for brain tumor detection analyzes the image. If tumors are detected above a confidence threshold, they are highlighted with bounding boxes on the image. The application displays both the original and the processed image side-by-side and indicates if no tumors were detected. Users can also download the processed image.
- Image Upload: Supports PNG, JPG, and JPEG formats.
- Tumor Detection: Utilizes a YOLOv11 model to identify potential tumors.
- Visual Feedback: Displays the original image and the processed image with bounding boxes highlighting detected tumors.
- Clear Results: Explicitly states "No cancer detected" if the model finds no tumors above the confidence threshold.
- Download Results: Allows users to download the processed image with annotations.
- User-Friendly Interface: Built with Streamlit for simple interaction.
- Image Resizing: Automatically resizes large images to prevent memory issues while maintaining aspect ratio.