pottery_utilities.py¶

Overview¶

This file contains 28 documented elements.

Classes¶

PDFExtractor¶

Extract images from PDF files and image files - Step 1 of PyPotteryLens workflow

Methods¶

init(self)¶

Initializes a new PDFExtractor instance with default configuration values. Sets supported_formats to a predefined list of image file extensions, and initializes yolo_model and model_loaded to None and False respectively, indicating no model is loaded at construction time. An empty dictionary mask_outputs is also initialized to store masks for later extraction.

extract_from_images(self, image_paths, output_dir, auto_detect, confidence, kernel_size, iterations)¶

Extract pottery drawings from image files (JPEG, PNG, etc.)

Args: image_paths: List of image file paths or single path output_dir: Output directory for processed images auto_detect: Auto-detect pottery using YOLO model confidence: Detection confidence threshold kernel_size: Size of morphological kernel iterations: Number of dilation iterations

Returns: List of extracted/processed image paths

load_yolo_model(self)¶

Load YOLO model for pottery detection if available

detect_and_extract_pottery(self, image_path, confidence, kernel_size, iterations)¶

Detect pottery in image and extract with morphological operations

Args: image_path: Path to image confidence: Detection confidence threshold kernel_size: Size of morphological kernel iterations: Number of dilation iterations

Returns: List of detected pottery regions with masks

extract(self, pdf_path, output_dir, split_pages, auto_detect, confidence, kernel_size, iterations)¶

Extract images from PDF

Args: pdf_path: Path to PDF file output_dir: Output directory for images split_pages: Split pages into left/right auto_detect: Auto-detect pottery images using YOLO confidence: Detection confidence threshold

Returns: List of extracted image paths

LayoutGenerator¶

Generate image layouts for pottery documentation

Methods¶

init(self)¶

No description available. Initializes a new LayoutGenerator instance. Sets up the font_cache attribute as an empty dictionary, which is used to store fonts for text rendering.

get_font(self, size)¶

Get font for text rendering

create_preview(self, images, mode, page_size, rows, cols)¶

Create a preview of the layout

generate(self, images, output_path, mode, page_size, rows, cols, add_captions, add_scale, scale_cm)¶

Generate the final layout

Args: images: List of image paths output_path: Output file path mode: Layout mode ('grid' or 'puzzle') page_size: Page size rows: Number of rows (for grid mode) cols: Number of columns (for grid mode) add_captions: Add image captions add_scale: Add scale bar scale_cm: Scale bar length in cm

Returns: True if successful

ImageProcessor¶

Process and enhance pottery images

Methods¶

init(self)¶

No description available. Initializes a new instance of the ImageProcessor class. No parameters are accepted and no initialization logic is performed beyond object instantiation.

enhance_image(self, image_path, output_path, brightness, contrast, sharpness)¶

Enhance image quality

Args: image_path: Input image path output_path: Output image path brightness: Brightness adjustment (1.0 = no change) contrast: Contrast adjustment (1.0 = no change) sharpness: Sharpness adjustment (1.0 = no change)

Returns: True if successful

remove_background(self, image_path, output_path, threshold)¶

Simple background removal (white background)

Args: image_path: Input image path output_path: Output image path threshold: Threshold for white detection

Returns: True if successful

PotteryInkProcessor¶

AI-powered pottery drawing enhancement using diffusion models Transforms pencil drawings into publication-ready illustrations Based on PyPotteryInk project

Methods¶

init(self, venv_python)¶

Initializes a new instance of the pencil-drawing-to-illustration transformer, setting model, device, use_fp16, model_cache, and venv_python to their default values. The venv_python parameter accepts an optional path to a virtual environment Python executable used for subprocess calls. After initialization, _setup_device is called automatically to detect and configure the best available processing device.

is_available(self)¶

Check if PotteryInk functionality is available

load_model(self, model_path)¶

Load PotteryInk model for processing

enhance_drawing(self, input_image_path, output_path, prompt, contrast_scale, patch_size, overlap, apply_preprocessing, model_path, use_ml, manual_contrast, manual_brightness, background_mode, bg_threshold)¶

Transform a pencil drawing into publication-ready illustration

Args: input_image_path: Path to input pencil drawing output_path: Path for output illustration prompt: Processing prompt for AI model contrast_scale: Contrast enhancement factor (also used as strength for ML) patch_size: Size of processing patches overlap: Overlap between patches apply_preprocessing: Whether to apply preprocessing model_path: Path to .pkl model file (optional, will use default if not provided) use_ml: Whether to use ML-based enhancement (True) or basic filters (False) manual_contrast: Manual contrast adjustment (1.0 = no change) manual_brightness: Manual brightness adjustment (1.0 = no change) background_mode: Background treatment ("Keep Original", "White", "Transparent") bg_threshold: Brightness threshold for background detection (200-255)

Returns: True if successful, False otherwise

extract_elements(self, image_path, output_dir, min_area)¶

Extract individual pottery elements from processed image

Args: image_path: Path to processed image output_dir: Directory to save extracted elements min_area: Minimum area for detected elements

Returns: List of paths to extracted element images

batch_process(self, input_folder, output_folder, model_path, prompt, progress_callback)¶

Process multiple drawings in batch

Args: input_folder: Folder containing input drawings output_folder: Folder for processed outputs model_path: Path to PotteryInk model prompt: Processing prompt progress_callback: Function to call for progress updates

Returns: Dictionary with processing results and statistics

enhance_high_res(self, input_image_path, output_path, patch_size, overlap, contrast_scale, apply_preprocessing, manual_contrast, manual_brightness, background_mode, bg_threshold)¶

Process high-resolution images using patch-based processing

Args: input_image_path: Path to input image output_path: Path for output image patch_size: Size of processing patches overlap: Overlap between patches in pixels contrast_scale: Contrast enhancement factor apply_preprocessing: Whether to apply preprocessing manual_contrast: Manual contrast adjustment (1.0 = no change) manual_brightness: Manual brightness adjustment (1.0 = no change) background_mode: Background treatment ("Keep Original", "White", "Transparent") bg_threshold: Brightness threshold for background detection (200-255)

Returns: True if successful, False otherwise

export_to_svg(self, image_path, output_path, simplify, min_path_size)¶

Export enhanced pottery drawing to SVG format

Args: image_path: Path to enhanced image output_path: Path for output SVG file simplify: Simplification factor for paths min_path_size: Minimum size for paths to include

Returns: True if successful, False otherwise

apply_preprocessing(self, image, model_stats)¶

Apply recommended preprocessing adjustments based on model statistics

Args: image: Input PIL Image model_stats: Model statistics dictionary (if None, uses default)

Returns: Preprocessed PIL Image

analyze_dataset_metrics(self, folder_path)¶

Analyze all images in a folder to extract statistical metrics

Args: folder_path: Path to folder containing images

Returns: Dictionary with dataset statistics

run_diagnostic(self, image_path)¶

Run diagnostic analysis on pottery drawing

Args: image_path: Path to image to analyze

Returns: Dictionary with diagnostic information