README.md

pixelshop

pixelshop is a python fluent interface wrapper for image manipulation using the open-source Computer Vision library OpenCV.

This approach allows method chaining, making the code more legible. To avoid unwanted side effects a deep copy is created before each method call.

Warning

pixelshop is in development.

Examples

Example 1: Find bon silhoutte

#!/usr/bin/env python3

from pixelshop.flags import ThresholdTypes
from pixelshop.image import DevelopedImage

bon = (
    DevelopedImage.fromFile("examples/bon.jpg", proxyPixelCount=500 * 500)
    .useProxy()  # Proceed with a low resolution version to improve speed
    .appendToFilename("small")
    .saveImage()
    .colorBGR2HSV()  # Convert to HSV
    .keepChannels(keepChannels=[1, 3, 3])  # Keep the average of channels 1, 3 and 3
    .denoise(strength=15)  # Denoise
    .morphologicalTransformClosing(kernelSize=(3, 3), iterations=2)  # Remove text
    .adaptiveThreshold(
        thresholdType=ThresholdTypes.THRESH_BINARY,
        blockSize=1000,
        C=-12,
    )  # Threshold the image to get silhouette
    .morphologicalTransformClosing(kernelSize=(3, 3), iterations=3)  # Final clean up
    .appendToFilename("silhouette")
    .saveImage()
    .show()  # Display result
)

Code file

original	silhouette

Example 2: Enhance sudoku grid

#!/usr/bin/env python3

from pixelshop.flags import ThresholdTypes
from pixelshop.image import DevelopedImage

bon = (
    DevelopedImage.fromFile("examples/sudoku.png", proxyPixelCount=500 * 500)
    .useProxy()  # Proceed with a low resolution version to improve speed
    .appendToFilename("small")
    .saveImage()
    .colorBGR2Gray()  # Convert to gray
    .show()
    .denoise(strength=3)
    .show()
    .adaptiveThreshold(
        thresholdType=ThresholdTypes.THRESH_BINARY, blockSize=250
    )  # Threshold the image to get grid
    .appendToFilename("enhanced")
    .saveImage()
    .show()  # Display result
)

Code file

original	silhouette

Installation

A. Install package manager uv

1. Install uv by following Astral's Getting Started guide.
   • macOS: using Homebrew with brew install uv

B. Install pixelshop

1. Clone the repository using git clone https://github.com/davidclemens/pixelshop.git <local-path-to-clone-to>.

2. Create a virtual python environment

   1. cd into the directory of the repository (<local-path-to-clone-to> above)

   2. Create a virtual python 3.12 environment with

      uv venv -p 3.12

3. Install Python dependencies

   1. Activate the virtual environment with

      source .venv/bin/activate

   2. Install the requirements with

      uv pip install -r requirements.txt

Architecture

Method chaining

The method chaining without side effects is implemented in the @chainable decorator.

Dependencies

• OpenCV for image processing: source, docs, website
• NumPy: source, docs, website