PyImageKit

A Python library providing a unified interface for handling images across multiple formats (bytes, OpenCV cv2, Pillow Image) and performing common image processing tasks.

Installation

pip install pyimagekit

To include OCR support for document uprighting, install with the optional dependency:

pip install "pyimagekit[pytesseract]"

Quick Start

pyimagekit centers around the SmartImage class, which allows you to seamlessly load, transform, and export images. It implements a fluent interface so operations can be chained.

from pyimagekit import SmartImage

# 1. Load an image
with open("document.jpg", "rb") as f:
    img = SmartImage.from_bytes(f.read())

# Alternatively:
# img = SmartImage.from_cv2(cv2_image)
# img = SmartImage.from_pillow(pil_image)

# 2. Process the image
processed_img = (
    img.resize(width=800)
       .rotate(90)
       .ensure_document_portrait()
)

# 3. Export the processed image
jpeg_bytes = processed_img.to_bytes(format="JPEG", quality=85)
base64_str = processed_img.to_base64(format="PNG")
cv2_mat = processed_img.to_cv2()
pillow_img = processed_img.to_pillow()

Features

The SmartImage class currently supports the following operations:

• Loading: from_bytes, from_cv2, from_pillow
• Exporting: to_cv2, to_pillow, to_bytes, to_base64
• Transformations: resize, crop, rotate, sharpen_v1, sharpen_v2, adjust_quality, apply (for custom functions)
• Document Processing: unwarp_document, ensure_document_upright, ensure_document_portrait, ensure_document_landscape

License

MIT License