peafowl-dox 0.1.0

No project description provided

Peafowl Dox

A utility library for image and document processing. Essential tools for handling multipart uploads, PDF conversion, and preparing documents for OCR and ML pipelines.

Table of Contents

• Installation
• Quick Start
• API Reference
  • Image Upload Processing
  • PDF Conversion
  • Image Resizing
  • OCR Preparation
  • Image Processor Class
• Error Handling
• Depencies

---

Installation

pip install peafowl_dox

---

Quick Start

from fastapi import UploadFile
from peafowl_dox import multipart_to_array

@app.post("/upload/")
async def upload_image(file: UploadFile):
        image_array = multipart_to_array(file.file)
        print(f"Image shape: {image_array.shape}")
        return {"message": "Image processed successfully"}

---

API Reference

Image Upload Processing

Convert multipart file uploads to numpy arrays.

from peafowl_dox import multipart_to_array

# From FastAPI upload
array = multipart_to_array(file.file)

# From Flask upload
array = multipart_to_array(request.files['image'])

# From BytesIO
from io import BytesIO
with open('image.jpg', 'rb') as f:
        buffer = BytesIO(f.read())
        array = multipart_to_array(buffer)

Returns: np.ndarray with shape (height, width, channels)

---

PDF Conversion

Convert PDF pages to image arrays.

from peafowl_dox import pdf_to_images

# From file path
images = pdf_to_images("document.pdf", dpi=150)

# From bytes
with open("document.pdf", "rb") as f:
        images = pdf_to_images(f.read(), dpi=200)

# From upload
images = pdf_to_images(file.file, dpi=150)

print(f"Converted {len(images)} pages")

Parameters:

• pdf_input: File path, bytes, or file-like object
• dpi: Resolution (default: 150)
• image_format: "RGB", "RGBA", or "L" (grayscale)

Returns: List of numpy arrays (one per page)

---

Image Resizing

Resize images with aspect ratio preservation.

from peafowl_dox import resize_image

# Resize to specific dimensions
resized = resize_image(image, (800, 600), maintain_aspect=True)

# Resize by max dimension
resized = resize_image(image, 1024)  # Max 1024px

# Resize without aspect ratio
resized = resize_image(image, (800, 600), maintain_aspect=False)

Parameters:

• image: Numpy array
• target_size: (width, height) tuple or single int for max dimension
• maintain_aspect: Preserve aspect ratio (default: True)
• interpolation: OpenCV interpolation method (default: cv2.INTER_AREA)

---

OCR Preparation

Prepare images for OCR with enhancement.

from peafowl_dox import prepare_for_ocr

# Basic preparation
ocr_ready = prepare_for_ocr(image)

# With resize and enhancement
ocr_ready = prepare_for_ocr(image, target_size=1200, enhance=True)

# Custom size without enhancement
ocr_ready = prepare_for_ocr(image, target_size=(1024, 768), enhance=False)

Parameters:

• image: Numpy array
• target_size: Optional resize target (int or tuple)
• enhance: Apply noise reduction and contrast enhancement (default: True)

Returns: Grayscale numpy array optimized for OCR

---

Image Processor Class

Full-featured processor for complex workflows.

from peafowl_dox import ImageProcessor

processor = ImageProcessor(default_ocr_size=1200)

# Process from path
image = processor.process_image("path/to/image.jpg")

# Process from bytes
with open("image.jpg", "rb") as f:
        image = processor.process_image(f.read())

# Process from numpy array
image = processor.process_image(existing_array)

# Process directly for OCR
ocr_ready = processor.process_for_ocr("path/to/scan.jpg", enhance=True)

Methods:

• process_image(img): Accepts path, bytes, or numpy array → returns RGB array
• process_for_ocr(img, target_size, enhance): Process and prepare for OCR

---

Error Handling

from peafowl_dox import (
        PeafowlDoxError,
        ImageProcessingError,
        PDFConversionError
)

try:
        images = pdf_to_images("document.pdf")
except PDFConversionError as e:
        print(f"PDF conversion failed: {e}")

try:
        array = multipart_to_array(file)
except ImageProcessingError as e:
        print(f"Image processing failed: {e}")

---

Dependencies

• Python >= 3.8
• numpy >= 1.24.0
• Pillow >= 10.0.0
• opencv-python >= 4.8.0
• PyMuPDF >= 1.23.0

---

Changelog

[0.1.0] - 2025-11-06

• Initial release
• multipart_to_array: Convert uploads to numpy arrays
• pdf_to_images: PDF to image conversion
• resize_image: Smart image resizing
• prepare_for_ocr: OCR preparation with enhancement
• ImageProcessor: Full-featured processing class