Image Processing Application - DIP Midterm Project

This is a image processing application I made for Digital Image Processing course. I used PyQt5 to make a simple GUI where you can load images and do different image processing operations. The source code files are in src folder and executable file is in dist folder.

App Overview

The application has two parts. Left side has control panel with operation buttons. Right side has two image panels showing original and result images.

Implemented Features

File Operations: Load Image, Save Image, Reset

Basic Operations: Grayscale, Flip Horizontal/Vertical, Rotate +90/-90 degrees

Affine Transformations: Rotate (custom angle), Scale (uniform or X/Y), Translate (dx, dy), Shear (X/Y/Both)

Intensity Adjustments: Contrast (slider 50-200%), Contrast Stretch, Gamma (slider 0.1-3.0), Negative

Spatial Filters: Mean/Box (5x5), Gaussian (15x15), Median (5x5), Laplacian (3x3), Sobel X/Y (3x3)

Histogram Operations: Show Histogram, Histogram Equalization

Morphological Operations: Global threshold (0-255), Otsu threshold (automatic), Erode, Dilate, Open, Close

Key Parameters

Threshold: Global uses manual value (0-255, default 127). Otsu is automatic.

Morphological Operations: Uses 3x3 ellipse kernel, 1 iteration. Need to apply threshold first.

Spatial Filters: Kernel sizes are fixed (Mean 5x5, Gaussian 15x15, Median 5x5, Laplacian/Sobel 3x3). All convert to grayscale.

Scale: Positive numbers only (0.5 = half size, 2.0 = double size).

Contrast: Slider 50-200% (100% = no change).

Gamma: Slider 0.1-3.0 (1.0 = no change).

How to Use

1. Click Load Image to select file
2. Click operation button from left panel
3. See result in right panel
4. For morphology: Apply Threshold first, then use Erode/Dilate/Open/Close
5. Click Save Image to save result
6. Click Reset to go back to original

Known Limitations

• Filters and morphology work on grayscale only (color images converted automatically)
• Large images might be slow
• Scale must be positive number
• Morphology needs threshold first
• Large scale values (over 10x) might cause problems

Installation

pip install -r requirements.txt
python app_gui_qt.py

Requirements

Python 3.9+, PyQt5, OpenCV, NumPy, Pillow

Course: Digital Image Processing (DIP) - Midterm Project