AI Image Screener

A practical first-pass AI image screening system designed to identify images that require human review based on statistical and physical patterns.

Screening accuracy: 40-90% detection rate across AI models

Fast Processing

Parallel processing for batch analysis with real-time progress tracking

Multi-Signal Detection

Five independent statistical detectors with weighted ensemble aggregation

Comprehensive Reports

Export results in CSV, JSON, and PDF formats for integration and documentation

Adjustable Sensitivity

Conservative, balanced, and aggressive modes for different use cases

Important Notice

This is not a perfect AI detector. It's a screening tool that helps reduce manual review workload by flagging suspicious images for human verification.

Gradient-Field PCA
Weight: 30%

Detects lighting & gradient inconsistencies typical of diffusion models. Analyzes directional light patterns and shadow consistency that often appear unnatural in AI-generated images.

Detection Method Principal Component Analysis
Sensitivity High for diffusion models
Performance 85-95% detection rate
Frequency Analysis
Weight: 25%

Identifies unnatural spectral energy distributions via FFT analysis. AI-generated images often show characteristic frequency patterns different from camera-captured photos.

Detection Method Fast Fourier Transform
Sensitivity Medium-High
Performance 75-85% detection rate
Noise Pattern Analysis
Weight: 20%

Detects missing or artificial sensor noise patterns. Real cameras produce characteristic noise while AI models often generate unnaturally uniform or missing noise patterns.

Detection Method Noise Distribution Analysis
Sensitivity Medium
Performance 70-80% detection rate
Texture Statistics
Weight: 15%

Identifies overly smooth or uniform texture regions. AI-generated images often lack the natural texture variation found in real photographs, especially in complex surfaces.

Detection Method GLCM Texture Analysis
Sensitivity Medium-Low
Performance 60-70% detection rate
Color Distribution
Weight: 10%

Flags unnatural saturation and color histogram patterns. AI models often produce colors that are either oversaturated or have distribution patterns that differ from real photographs.

Detection Method Color Histogram Analysis
Sensitivity Low-Medium
Performance 50-65% detection rate
1

Upload Images

Drag & drop or select images (JPG, PNG, WEBP)

2

Start Analysis

Click "Start Analysis" to begin screening

3

Review Results

Check flagged images and export reports