Command Line Interface

ImgVisFeat provides a command-line interface (CLI) built with Typer for easy access to its visualization features.

Basic Usage

imvf [COMMAND] IMAGE_PATH

The CLI uses subcommands for different visualization methods, making it intuitive and easy to use.

Getting Help

# Show main help with all available commands
imvf --help

# Show help for a specific command
imvf all --help
imvf hog --help

Available Commands

`all` - Visualize All Features

Applies all visualization methods to the input image.

imvf all path/to/image.jpg

`color-channel` - Color Channel Visualization

Extracts and visualizes individual RGB color channels.

imvf color-channel path/to/image.jpg

`gradient` - Gradient Visualization

Computes and visualizes image gradients in X, Y, and combined XY directions.

imvf gradient path/to/image.jpg

`hog` - Histogram of Oriented Gradients

Visualizes HoG feature descriptors for object detection.

imvf hog path/to/image.jpg

`keypoint` - Keypoint Detection

Detects and visualizes keypoints using SIFT, AKAZE, and ORB algorithms.

imvf keypoint path/to/image.jpg

`lbp` - Local Binary Patterns

Extracts and visualizes LBP texture descriptors.

imvf lbp path/to/image.jpg

`power-spectrum` - Power Spectrum Analysis

Analyzes and visualizes frequency domain characteristics.

imvf power-spectrum path/to/image.jpg

Examples

Basic Examples

# Visualize all features
imvf all input_image.jpg

# Visualize only HoG features
imvf hog input_image.jpg

# Visualize keypoints
imvf keypoint input_image.jpg

# Visualize color channels
imvf color-channel input_image.jpg

Using with Different Image Formats

# Works with various image formats
imvf all image.jpg
imvf gradient image.png
imvf hog image.bmp

Features

Built with Typer

The CLI is built with Typer, providing:

Beautiful help messages with formatted output
Shell completion support (install with imvf --install-completion)
Type-safe arguments with automatic validation
Intuitive subcommand structure

Auto-completion

Install shell completion for enhanced productivity:

# Install completion for your shell
imvf --install-completion

# Show completion script
imvf --show-completion

Implementation Details

The CLI is implemented in the imvf.cli module using the Typer framework.

imvf.cli.app `module-attribute`

app = Typer(help='ImgVisFeat: Visualize image features')

Error Handling

If an error occurs during the visualization process:

Error messages are printed to stderr
The command exits with code 1
A helpful error message indicates what went wrong

Example error output:

$ imvf all nonexistent.jpg
Error: Image not found: nonexistent.jpg
Visualization failed.

Output

All visualization methods:

Display results in OpenCV windows
Save results to a directory named after the input image
Print progress information to stdout

Example output:

$ imvf all sample.jpg
Visualizing sample.jpg
Number of keypoints (SIFT) : 957
Number of keypoints (AKAZE): 513
Number of keypoints (ORB): 500
Visualization complete.

Notes

Each subcommand represents a specific visualization method
The all command runs all available visualization methods
Results are automatically saved to an output directory
Press any key in the OpenCV window to close the visualization