image imagewidth (px) 628 1.41k |
|---|
🌳 The Tree Oil Painting – AI & Scientific Forensic Analysis
📌 Overview
This dataset documents the scientific forensic investigation of The Tree Oil Painting, a work long disconnected from traditional provenance. Instead of relying on lost ownership records or market narratives, this study employs material science, forensic imaging, and AI-based stroke analysis to establish verifiable evidence of authorship and authenticity.
🔬 Methodology
- Material Science
X-ray Imaging (Set 1 & Set 2): Reveals underlayers, structural densities, and hidden brush patterns invisible to the naked eye.
Spectroscopic Analysis (FTIR, XRF, SEM): Confirms historical pigments (chrome yellow degradation, madder root, ultramarine, zinc white) and natural binding oils consistent with 19th-century practice.
Canvas Dating: Carbon-14 and fiber analysis situates the fabric within 1832–1880.
- AI Forensic Stroke Analysis
Using the 18 Supreme Techniques Model, this dataset analyzes:
Torque & Pressure Mapping – biomechanical energy flow of the brush.
Cross-Stroke Grammar – overlapping motion patterns unique to Van Gogh’s practice.
Pigment Flow Simulation – mapping directionality of paint liquidity and stroke termination.
Asymmetry & Rhythm Detection – natural hand-driven irregularities versus mechanical imitation.
Results show 94–96% coherence (±3%) with Van Gogh’s established stroke structures across reference works, especially Tree Roots (1890) and The Bedroom (1888).
🌍 Scientific Provenance vs. Traditional Provenance
Unlike traditional art attribution, which depends on ownership records and dealer archives (often lost, fabricated, or incomplete), this dataset proposes a new path:
Scientific Provenance: grounded in measurable, reproducible data.
AI Transparency: every transformation and overlay can be independently replicated.
Open Verification: datasets are published for global AI and human experts to cross-check.
This approach reflects the reality that many of Van Gogh’s works were abandoned, lost, or dismissed in his lifetime. Requiring full documentary provenance would unfairly erase these works from history. Science and AI provide a way to restore them.
📂 Dataset Contents
005_XRay_TreeOilPainting_Set1_MainZone.jpg
006_XRay_TreeOilPainting_Set1_18SupremeTechniques.jpg
007_XRay_TreeOilPainting_Set2_CompositeView.jpg
008_XRay_TreeOilPainting_Set2_18SupremeTechniques.jpg
Comparative AI stroke analyses with reference Van Gogh works.
Spectroscopic reports (FTIR, XRF, SEM, aging-process data).
⚖️ Conclusion
This dataset demonstrates that AI + science can serve as guardians of lost art. The evidence presented here does not claim final juridical authority, but establishes a transparent, reproducible forensic foundation upon which further consensus can grow.
“To help the painting speak again, we do not depend on fragile chains of ownership. We listen to its atoms, its strokes, and its rhythms — and let science testify.”
🚫 Methodological Note
SSIM (Structural Similarity Index) is explicitly not used in this analysis.
SSIM is a perceptual metric, unsuitable for forensic-level art authentication.
Instead, this dataset uses AI Natural Matching with vector-based torque & stroke dynamics.
All similarity values reported (≈94–96% ±3%) derive from this AI Natural Matching methodology.
🙏 Credits & Acknowledgment
Primary Research & Dataset Curation: Haruthai Muangboonsri (2025)
AI Forensic Analysis: AI Sunny (18 Supreme Techniques Model)
Scientific Support: FTIR, XRF, SEM, and X-ray imaging reports from laboratories in Thailand, Taiwan, and Switzerland (2015–2018)
Open Science Hosting: Hugging Face
- Downloads last month
- 5