Image Classifier

A lightweight Python-based desktop application for rapid image classification into user-defined categories. This tool allows users to quickly label images by assigning keyboard shortcuts to classes for fast classification.

Features

• Simple and intuitive user interface
• Custom class definition with keyboard shortcuts
• Single-key classification for rapid labeling
• Batch image processing from a selected folder
• Exports classifications to CSV format
• Supports common image formats (PNG, JPG, JPEG, GIF, BMP)

Installation

1. Clone this repository:

git clone https://github.com/yourusername/image-classifier.git
cd image-classifier

2. Install required dependencies:

pip install -r requirements.txt

Usage

1. Run the application:

python main.py

2. In the setup screen:

   • Enter a class name (e.g., "cat")
   • Enter a single-character shortcut key (e.g., "c")
   • Click "Add Class" to add it to the list
   • Repeat for all needed classes
   • You can remove classes from the list if needed
   • Click "Select Image Folder" to choose the folder containing your images
   • Click "Start Classification" to begin

3. In the classification screen:

   • Images will be displayed one at a time
   • Press the corresponding shortcut key to assign a class to the current image
   • The application will automatically move to the next image
   • Available classes and their shortcuts are displayed for reference

4. When finished:

   • A classifications.csv file will be created in the image folder
   • The CSV will contain two columns: filename and class name

Requirements

• Python 3.6 or higher
• Pillow (PIL Fork)
• tkinter (usually comes with Python)