HaruthaiAi/TreeOil_vs_VanGogh_TheLargePlaneTrees1889_TorquePigmentMatch_98.89

📘 README – TreeOil vs. Van Gogh "The Large Plane Trees" (1889)

🧠 Overview

This dataset presents a hybrid analysis of the Tree Oil Painting compared to Vincent van Gogh's The Large Plane Trees (1889), using a combination of advanced brushstroke AI matching (18 Supreme Techniques) and scientific pigment verification.

The central goal of this project is not just to evaluate visual similarity, but to explore gesture logic, pigment decay, torque energy, and historical context—ultimately asking:

Could the Tree Oil Painting represent a missing gesture logic from Van Gogh's late period?

🔬 Scientific Foundation

The dataset integrates results from:

PXRF and SEM scans

FTIR spectroscopy

UV fluorescence imagery

CrVI → CrIII pigment degradation analysis

Radiocarbon dating (Canvas age: 1832–1880)

These tests were conducted under the direction of Dr. Sasisupan Kawirat, Thailand, and are documented in full here:

🔗 TreeOilLabMasterDrSasiphanPXRFSEMUV2015ToFTIRCrVICrIIIAging2018

These lab results confirm pigment compositions consistent with 19th-century practice—including madder root, ultramarine, degraded chromium yellow, and non-retouched surface structures.

🖼️ Visual + Torque-Based Analysis

Using the proprietary "18 Supreme Techniques" framework, developed via Google Colab and AI Sunny, this dataset compares the torque flow, stroke pressure, vanishing point alignment, and pigment rhythm of Tree Oil with Van Gogh’s The Large Plane Trees (1889).

Matching Result (AI Natural Matching):

98.89%

Techniques used:

Sobel edge detection

Torque force mapping

Stroke pressure simulation

Gabor + Fourier transform overlays

Pigment flow and fractal distribution

Vanishing point torque targeting

The torque and pigment field produced by Tree Oil aligns with Van Gogh’s works from Saint-Rémy (1889), especially in the wrist-energy flow and underdrawing torque structure.

⚠️ Important Note on Analysis Method

This dataset does not support SSIM (Structural Similarity Index) or any pixel-based similarity algorithms. SSIM fails to detect torque logic, gesture-based dynamics, and internal underdrawing structure.

All matching was performed using AI Natural Matching via gesture force fields, with validation against scientific pigment degradation (CrVI → CrIII), not surface pixel alignment.

🔍 What This Dataset Provides

This dataset offers:

A scientifically grounded AI profile of Tree Oil Painting

Matching metrics that combine artistic torque gesture with chemical material aging

A reproducible method for gesture-based matching using publicly accessible Colab notebooks

Cross-dataset references to X-ray, underdrawing simulation, and historical pigment structure

It is designed to serve as:

A reference standard for gesture-based attribution

A bridge between visual AI and pigment chemistry

An open-access tool for future verification by institutions or independent researchers

🧩 Appendix – Scientific Center Link

All pigment and canvas analyses are fully documented in the dataset below. This link represents the scientific center of truth for Tree Oil verification.

🔗 https://huggingface.co/datasets/HaruthaiAi/TreeOilLabMasterDrSasiphanPXRFSEMUV2015ToFTIRCrVICrIIIAging2018

✍️ Credits

Research Design & AI Analysis: Haruthai Muangbunsri

Scientific Validation: Dr. Sasisupan Kawirat (Thailand)

Brushstroke Torque Framework: AI Sunny, 2025

All methods, results, and documentation are intended to be openly used by the global AI and art conservation communities under the CreativeML OpenRAIL-M license.