A modern, user-friendly web application for classifying images using a custom-trained Convolutional Neural Network (CNN).
Image Express is a fun and interactive web application designed to classify images of characters from two fictional universes: Pokémon and Thomas the Tank Engine.
The app uses a client-server architecture for smooth performance:
- The user selects the character set they want to classify.
- They upload an image from their local device through a clean SolidJS interface.
- The frontend sends the image to a Python backend server, which processes it and feeds it into a CNN model.
- The classification result is sent back to the frontend and displayed to the user.
This design ensures scalability, easy maintenance, and potential for future expansion.
- Category Selection: Users choose whether to classify a Pokémon or a Thomas the Tank Engine character.
- Interactive Image Upload: Drag-and-drop or file-picker interface for easy image upload (.jpg, .png, etc.).
- Real-Time Classification: Quick feedback displaying the classification result prominently.
- Image Classification CNN: Trained on extensive datasets for high-accuracy predictions.
- RESTful API: Enables seamless communication between SolidJS frontend and Python backend.
- Image Preprocessing: Automatic resizing, normalization, and formatting for CNN compatibility.
- Entertainment & Engagement: Fun app for fans to test or learn character names.
- Educational Extension: Can be extended as a learning tool for species, objects, or characters, or as the basis for a trivia game.
Built with: SolidJS (Frontend) | Python & CNN (Backend)