This is a desktop application that utilizes a Convolutional Neural Network (CNN) to classify images into defined categories. In this example, we are categorizing images of hamburgers and boxes.
The main libraries used in this project are Tkinter (for building the GUI), Keras (for building and training the CNN), and Scikit-learn (for data preprocessing and model evaluation).
The CNN model was trained on a dataset of images of hamburgers and boxes, and achieved high accuracy in distinguishing between the two classes
This application can be used in various scenarios such as in online orders where the user orders a product and receives a different product than what was specified. This implementation may not seem very useful (who would be interested in distinguishing a box from a hamburger?). But it builds the architecture for more sophisticated implementations.
This application allows users to:
- Load an image from their computer system
- View the image
- Rate the image
The main libraries used in this project are:
- Tkinter
- Keras
- scikit-learn
To run the application, simply run the gui_image_classifier.py file. This will launch a GUI where users can upload their images and classify them
Instantiate desktop app
Upload local image :
Run and Predict class :
