User Guide

This guide provides detailed information on using each visualizer in ImgVisFeat.

Available Visualizers

ImgVisFeat provides several specialized visualizers for different types of image analysis:

Feature Extraction

• Color Channels - Extract and visualize RGB color channels
• Gradients - Compute and visualize image gradients
• HoG Features - Histogram of Oriented Gradients for object detection
• LBP Features - Local Binary Patterns for texture analysis
• Keypoint Detection - Detect and visualize keypoints (SIFT, AKAZE, ORB)
• Power Spectrum - Frequency domain analysis

All-in-One

• Visualizer Class - Apply all visualizations at once

Common Usage Pattern

All individual visualizers follow the same usage pattern:

import cv2
import imvf

# Load image
image = cv2.imread("path/to/image.jpg")

# Create visualizer
visualizer = imvf.SomeVisualizer()

# Apply visualization
result = visualizer(image)

# Access results
# Results are dataclass instances with specific fields

Result Types

Each visualizer returns a typed result object containing the visualization results:

• ColorChannelResult - Blue, Green, Red channels
• GradientResult - Gradient X, Y, XY
• HogResult - HoG visualization
• LBPResult - LBP visualization
• KeypointResult - Keypoint and rich keypoint images
• PowerSpectrumResult - Power spectrum visualization

See the API Reference for complete details on each result type.