3% Weight - Color Science

Color Palette Analysis

Saturation & Color Diversity

Analyzes color distribution including saturation levels, color diversity, and white balance consistency. AI images often have oversaturated colors.

65-75%
Accuracy
3%
Weight
Color Palette Analysis

Frequently Asked Questions

What is color palette analysis?

Examination of color distribution including saturation levels, hue diversity, and color balance across the image to detect AI generation artifacts.

Why are AI images oversaturated?

AI models are often trained on heavily edited images with boosted colors. They learn to replicate this "enhanced" look, producing colors more vibrant than typical camera output.

What is color diversity?

Measured using hue histogram entropy, this indicates how many distinct colors are present. AI images may have artificially limited or exaggerated color variety.

How is white balance checked?

By comparing color temperature (red/blue ratio) across image regions. Real scenes have consistent temperature; AI may have unnatural variations.

Can post-processing fool this?

Yes, color grading can normalize AI colors. This is why color palette has low weight - it's a supporting indicator rather than definitive proof.

What saturation levels indicate AI?

Mean saturation above 140 (on 0-255 scale) suggests AI generation. Most camera photos fall in the 40-100 range without heavy editing.

Does this work on black and white photos?

Color palette analysis is not applicable to true B&W images. Other detection methods handle grayscale content more effectively.

What about stylized/artistic photos?

Heavily stylized real photos may trigger false positives. The low 3% weight prevents this from significantly affecting overall detection accuracy.

How is saturation uniformity analyzed?

Real photos have varying saturation based on lighting and materials. AI often produces unnaturally uniform saturation across different objects in a scene.

Can this detect specific AI models?

Different AI models have different color biases. Midjourney tends toward vibrant colors, while Stable Diffusion versions vary. The detector looks for general non-natural patterns.

Métodos relacionados

Detección ML

Nuestra detección ML usa modelos Transformer entrenados en millones de imágenes.

Análisis PRNU

Photo Response Non-Uniformity (PRNU) detects unique camera sensor fingerprints from manufacturing imperfections. AI images cannot replicate these authentic sensor signatures.

Análisis de Frecuencia

El análisis del dominio de frecuencia examina la distribución de componentes de alta y baja frecuencia en una imagen. Las imágenes generadas por IA típicamente carecen del ruido natural de alta frecuencia presente en fotografías reales.

Análisis de gradiente

Analyzes edge patterns and texture characteristics using Sobel, Canny, and Laplacian operators. AI images often have unnaturally smooth or uniform gradients.

Patrón de ruido

Las fotografías reales contienen patrones de ruido únicos de los sensores de cámara que varían a través de la imagen. Las imágenes generadas por IA tienen una distribución de ruido anormalmente uniforme.

Análisis de metadatos

Los metadatos de imagen contienen pistas valiosas sobre su origen. Analizamos datos EXIF, firmas de software y otra información incrustada para identificar herramientas de generación de IA.

Huella GAN

Detecta artefactos específicos de GAN como patrones de tablero de ajedrez y bandas de color.

Análisis de textura

Análisis Local Binary Pattern para anomalías de textura en imágenes IA.

Detección Anatómica

Los generadores de imágenes de IA a menudo crean errores anatómicos que los humanos reconocen inmediatamente como incorrectos. Usamos visión por computadora para detectar estos errores reveladores.

Verificación C2PA

C2PA (Coalition for Content Provenance and Authenticity) es un estándar de la industria para rastrear el origen y la historia del contenido digital a través de firmas criptográficas.

Semantic Inconsistency Detection

Detects logical inconsistencies like incorrect shadows, impossible perspectives, distorted reflections, and violations of physical laws that AI often produces.

Human Biometric Analysis

Uses MediaPipe to analyze human anatomy for incorrect finger counts, asymmetric eyes, unnatural skin texture, and other anatomical anomalies common in AI-generated faces.

Lighting Physics Validation

Validates light source consistency, shadow direction physics, specular highlight accuracy, and color temperature uniformity across the image.

Compression Artifact Analysis

Analyzes JPEG compression artifacts to estimate quality levels and detect re-compression patterns that indicate image manipulation or AI generation.

Edge Sharpness Analysis

Analyzes sharpness distribution across the image and validates depth-of-field consistency. AI often produces unnaturally uniform sharpness.

Statistical Pattern Analysis

Analyzes statistical properties including Shannon entropy, histogram patterns, and Benford's Law compliance to detect synthetic image characteristics.

Chromatic Aberration Analysis

Detects the absence of chromatic aberration (color fringing) that real camera lenses produce. AI images lack these optical artifacts.

Micro-Texture Analysis

Analyzes microscopic texture patterns for repetition, uniformity, and unnatural randomness that AI generators often exhibit.

Verificar Tu Imagen

Todos los métodos se combinan usando puntuación ponderada para producir un veredicto final con nivel de confianza.

Probar Ahora