Python API¶

Artefex provides a clean Python API for programmatic use.

Basic usage¶

from artefex import analyze, restore, grade

# Diagnose degradation chain
result = analyze("photo.jpg")
for d in result.degradations:
    print(f"{d.name}: {d.confidence:.0%} confidence, severity {d.severity:.0%}")

# Quality grade
grade_result = grade("photo.jpg")
print(f"Grade: {grade_result}")

# Restore
restore("photo.jpg", output="photo_restored.png")

Available functions¶

Function	Description
`analyze(path)`	Run all detectors, return `AnalysisResult`
`restore(path, output=None)`	Restore image, optionally save to output path
`grade(path)`	Return A-F quality grade
`compare(path_a, path_b)`	Compare two images (MSE, PSNR, SSIM)
`find_duplicates(directory)`	Find duplicate images in a directory
`generate_heatmap(path)`	Generate spatial degradation heatmap
`detect_platform(path)`	Identify source platform (Twitter, Instagram, etc.)

Data models¶

AnalysisResult¶

Returned by analyze(). Contains:

file_path - path to the analyzed file
file_size - size in bytes
dimensions - (width, height) tuple
format - image format string
degradations - list of Degradation objects

Degradation¶

Each detected degradation contains:

name - detector name (e.g. "jpeg_artifacts")
confidence - float 0.0-1.0
severity - float 0.0-1.0
detail - human-readable description
category - category string (e.g. "compression", "noise")

Batch processing¶

from artefex import analyze
from pathlib import Path

results = []
for img in Path("./photos").glob("*.jpg"):
    result = analyze(str(img))
    results.append(result)

# Sort by worst quality
results.sort(key=lambda r: len(r.degradations), reverse=True)

See examples/batch_processing.py for a more complete example.