Common Image Tools

Description

common_image_tools is a repository containing a curated collection of commonly used image manipulation functions for computer vision projects. These tools provide a set of handy functions for performing various image processing tasks, such as resizing, cropping, filtering, and more.

Features

• Resizing: Easily resize images to desired dimensions.
• Cropping: Crop images to focus on specific regions.
• Filtering: Apply various filters to enhance or modify images.
• Conversion: Convert images between different formats.
• Visualization: Tools for visualizing image data.
• Video Source: Flexible interface for handling different types of video inputs.

Installation

To install common_image_tools, run the following command:

pip install common-image-tools

or using poetry:

poetry add common-image-tools

Usage

Video Source

The VideoSource class provides a flexible interface for handling different types of video inputs with support for various OpenCV backends and hardware acceleration on Jetson devices.

Features

• Support for multiple video source types:
  • RTSP streams
  • Video files (MP4)
  • Webcam devices
• Configurable frame dimensions and FPS
• Multiple backend options:
  • OpenCV default
  • GStreamer
  • GStreamer with Jetson hardware acceleration
• Automatic backend selection based on system capabilities

Basic Usage

from common_image_tools import VideoSource
import cv2

# Create a video source
source = VideoSource(
    "video.mp4",
    target_frame_height=720,
    target_frame_width=1280,
    target_fps=30
)

# Use with OpenCV
cap = cv2.VideoCapture(source.parsed_source)

Different Source Types

# RTSP Stream
rtsp_source = VideoSource("rtsp://example.com/stream")

# Video File
file_source = VideoSource("path/to/video.mp4")

# Webcam
webcam_source = VideoSource("/dev/video0")

# Webcam
webcam_source = VideoSource(0)

Backend Selection

from common_image_tools import VideoSource, OpencvBackendMode

# Auto backend selection (default)
source = VideoSource("video.mp4")

# Force OpenCV default backend
source = VideoSource(
    "video.mp4",
    opencv_backend=OpencvBackendMode.OPENCV_DEFAULT
)

# Force GStreamer backend
source = VideoSource(
    "video.mp4",
    opencv_backend=OpencvBackendMode.OPENCV_GSTREAMER
)

# Use Jetson-optimized GStreamer backend
source = VideoSource(
    "video.mp4",
    opencv_backend=OpencvBackendMode.OPENCV_GSTREAMER_JETSON
)

API Reference

VideoSource

VideoSource(
    source,
    target_frame_height: int = None,
    target_frame_width: int = None,
    target_fps: int = None,
    opencv_backend: OpencvBackendMode = OpencvBackendMode.AUTO
)

Conversion

Operation

Tool

Verification

Visualization

Contributing

Contributions are welcome! If you have any suggestions, improvements, or new features to add, please open an issue or submit a pull request. Make sure to follow the existing coding style and include appropriate tests.

License

This project is licensed under the MIT License - see the LICENSE file for details.