Image Preprocessing Techniques

Python application implementing a two-phase image filter — histogram equalization and Gaussian convolution — with an interactive tkinter GUI. Simulates a mobile-style photo filter with real-time visualization and color style switching.

---

Overview

This project demonstrates core image processing concepts applied to a practical use case: a "mobile camera filter" pipeline. Users can load an image, apply preprocessing techniques sequentially, and compare results interactively.

---

Tech Stack

Tool	Purpose
Python	Core language
OpenCV	Image processing (convolution, histogram)
NumPy	Matrix operations
tkinter	Graphical user interface
Matplotlib	Result visualization and color maps

---

Project Structure

Img-preprocessing-techniques/
├── interfaz.py               # Main entry point — launches the GUI
├── hist.py                   # Histogram equalization module (standalone test)
├── convolucion_proyecto.py   # Gaussian convolution module (standalone test)
├── images/
│   ├── original_equalized.png
│   └── interfaz.png
└── README.md

---

Pipeline

Phase 1 — Histogram Equalization

Improves image contrast by redistributing pixel intensity values across the histogram range. This enhances fine details that may be lost in low-contrast images.

Original image → Spread intensity distribution → Improved contrast

Phase 2 — Gaussian Convolution (Blur Filter)

Applies a Gaussian kernel via convolution to smooth the image. Multiple kernel sizes are available for testing. A cmap parameter allows applying different color styles post-processing.

Equalized image → Gaussian kernel convolution → Styled output

---

GUI Usage

Run the main interface:

python interfaz.py

Workflow:

1. Enter the image filename in the text box
2. Click Start to load and display the image
3. Click Equalize to apply histogram equalization
4. Click Apply Gaussian Filter to run the convolution
5. Select one of 3 color styles to apply a visual effect

---

Running Modules Independently

# Test histogram equalization only
python hist.py

# Test Gaussian convolution only
python convolucion_proyecto.py

---

Installation

pip install opencv-python numpy matplotlib

tkinter is included in the Python standard library.

---

Concepts Demonstrated

• Histogram equalization for contrast enhancement
• 2D convolution with Gaussian kernels
• Real-time image processing pipeline
• GUI design with tkinter
• Matplotlib color maps for visual styling

---

Authors

Kevin Joan Delgado Pérez - Tecnológico de Monterrey