An Experiment Project with API and Web Interface created to Find Images Within Another Image
The Following technologies are used in this project
- openCV
- numpy
- fastapi
- streamlit
This is the UI - where you can upload your images

This is the Result - You can see the output highlighted in green annotation - Rectangle

The logical comparision is now powered by OpenCV's matchTemplate and cv2.TM_CCOEFF_NORMED
I am currently working on AI Custom Models to improve the accuracy of the results and to enhance the scope of the project
Here are some modeles I am trying right now
- Yolov8
- CNN Model
- OpenCV DNN Model https://docs.opencv.org/4.x/d2/d58/tutorial_table_of_content_dnn.html
cv2.matchTemplate(graybaseImage, scaled_refImg, cv2.TM_CCOEFF_NORMED)
This is an experiment project created to Find Images Within Another Image and to test the performance of OpenCV's
matchTemplateandcv2.TM_CCOEFF_NORMEDand other algorithms and provide an interface for future AI - ML based model validations - Feel free to contribute
- Download the project
git clone https://github.com/robertoferreira/image-within.git
- Install the requirements
pip install -r requirements.txt
- Run the UI
streamlit run ui/app.py
- Run the API
cd api && uvicorn app:app --reload
- Open the browser and go to http://localhost:8501
- AI Custom Models
- More Algorithms
- More UI Improvements
MIT License