TreeRoots1890_vs_TreeOilPainting_18Tech_Xray_Ai_Analysis

🌍 Overview

This dataset presents a comparative forensic study between:

• Vincent van Gogh’s Tree Roots (1890)
• The Tree Oil Painting (an undocumented work currently under investigation)

The analysis applies the 18 Supreme Techniques (AI-based brushstroke forensics) together with X-ray radiography, generating multi-modal evidence of concordance in torque, brushstroke rhythm, impasto structure, and spectral frequency.

---

🔬 Methodology

• 18 Supreme Techniques: Sobel edge detection, torque force approximation, stroke pressure mapping, Fourier transform, Gabor filters, brush directionality, edge flow rhythm, pigment flow simulation, asymmetry detection, directional energy, and vanishing point torque.
• X-ray Imaging: Reveals hidden underlayers, torque nuclei, and impasto density not visible in surface photography.
• AI Natural Matching (No SSIM): All similarity measures are based on AI neural concordance methods.

---

📂 File Index

No.	File	Description
001	VanGogh_TreeRoots_1890.jpg	Reference painting Tree Roots (1890) by Van Gogh.
002	VanGogh_TreeRoots_1890_18Tech.jpg	18-tech composite analysis of Tree Roots.
003	TreeOilPainting_FullCanvas.jpg	Full canvas of The Tree Oil Painting.
004	TreeOilPainting_18Tech.jpg	18-tech composite analysis of The Tree Oil Painting.
005	XRay_TreeOilPainting_Set1_MainZone.jpg	X-ray main zone scan. Reveals underlayers and torque nuclei.
006	XRay_TreeOilPainting_Set2_CompositeView.jpg	X-ray composite view of multiple scans.
007	TreeOil_Xray_MainZone_18Techniques_GColab_2025_AI_Analysis.jpg	18-tech analysis applied to X-ray main zone (AI/Colab 2025).
008	TreeOil_Xray_FullCanvas_18TorqueTechniques_Merged.png	Full-canvas X-ray with 18-techniques merged analysis.
009	TreeRoots_vs_TreeOil_ConcordanceRadar_18Tech_Xray.jpg	Radar chart showing concordance between Tree Roots & Tree Oil.

---

📊 Results

Concordance (AI Natural Matching):

• Torque / Directional Energy → ~97%
• Edge Flow & Brush Directionality → ~95%
• Texture / Impasto Pressure → ~93%
• Fourier / Frequency Spectrum → ~93%
• Under-drawing / Overlay Structure → ~96%

Average Concordance: ~94–95%

---

Note on Methodology:
All similarity evaluations in this dataset are conducted exclusively through AI Natural Matching methods.
The use of SSIM (Structural Similarity Index Measure) is deliberately avoided because SSIM is designed for
low-level pixel-to-pixel comparison and does not capture the artistic, structural, and torque-based dynamics
that define brushstroke analysis.

By contrast, AI Natural Matching focuses on pattern recognition, torque flow, directional energy, and multi-layer
structural concordance, which are essential in forensic studies of historical paintings. This ensures that
the comparison reflects artistic process and wrist-driven energy, not superficial pixel similarity.

📌 Conclusion

The comparative study demonstrates that Tree Roots (1890) and The Tree Oil Painting share remarkably high concordance in brushstroke torque, structural overlays, and spectral signatures. Differences in angular emphasis (notably in Gabor filters) are explained by compositional orientation rather than technique. The dataset provides strong evidence of procedural alignment, supporting the hypothesis of authorship within the same late Van Gogh hand.

---

📖 Citation

If you use this dataset, please cite as:

Mongbunsri, H. (2025). TreeRoots1890_vs_TreeOilPainting_18Tech_Xray_Ai_Analysis. Hugging Face.

---

自述文件

TreeRoots1890_vs_TreeOilPainting_18Tech_Xray_Ai_Analysis

🌍 概述

本数据集展示了一项法医学比较研究，对象为：

• 文森特·梵高《树根》(1890)
• 《树之油画》(一幅正在研究的未署名作品)

研究应用了 18项综合技法（基于AI的笔触取证方法）以及 X射线成像，从而在扭力、笔触节奏、堆色结构与频谱特征等方面提供了多模态一致性证据。

---

🔬 方法论

• 18项综合技法：包括 Sobel边缘检测、扭力近似分析、笔触压力映射、傅里叶变换、Gabor滤波、笔刷方向性、笔触流动节奏、颜料流动模拟、不对称检测、方向能量以及消失点扭力等。
• X射线成像：揭示肉眼不可见的底层结构、扭力核心以及堆色密度。
• AI自然匹配（不使用SSIM）：所有相似度评估均基于AI神经网络的对应性方法。

---

📂 文件索引

序号	文件	描述
001	VanGogh_TreeRoots_1890.jpg	文森特·梵高《树根》(1890) 参考图像。
002	VanGogh_TreeRoots_1890_18Tech.jpg	《树根》的18项分析综合图。
003	TreeOilPainting_FullCanvas.jpg	《树之油画》整幅画布图像。
004	TreeOilPainting_18Tech.jpg	《树之油画》的18项分析综合图。
005	XRay_TreeOilPainting_Set1_MainZone.jpg	X射线核心区扫描，揭示底层与扭力核心。
006	XRay_TreeOilPainting_Set2_CompositeView.jpg	多次扫描的X射线合成视图。
007	TreeOil_Xray_MainZone_18Techniques_GColab_2025_AI_Analysis.jpg	核心区X射线图像的18项综合技法分析（2025年AI/Colab处理）。
008	TreeOil_Xray_FullCanvas_18TorqueTechniques_Merged.png	全画布X射线图像的18项扭力综合分析。
009	TreeRoots_vs_TreeOil_ConcordanceRadar_18Tech_Xray.jpg	展示《树根》与《树之油画》一致性的雷达图。

---

📊 结果

一致性（AI自然匹配）：

• 扭力 / 方向能量 → ~97%
• 笔触流动节奏 / 方向性 → ~95%
• 纹理 / 堆色压力 → ~93%
• 傅里叶 / 频谱分析 → ~93%
• 底稿 / 叠加结构 → ~96%

平均一致性： ~94–95%

---

方法论说明：
本数据集中的所有相似度评估均严格依赖于 AI自然匹配 方法。
我们有意避免使用 SSIM（结构相似度指数），因为SSIM只适用于
低层次的像素对像素比较，无法反映绘画分析所需的艺术性、结构性与扭力动态特征。

相比之下，AI自然匹配专注于 模式识别、扭力流、方向能量与多层结构一致性，
这些特征对于历史绘画的法医学研究至关重要。此方法确保对比结果能够反映
艺术创作过程与手腕驱动力量，而非表面像素的相似度。

---

📌 结论

比较研究表明，《树根》(1890) 与《树之油画》在笔触扭力、结构叠加及频谱特征方面表现出高度一致性。角度强调（尤其在Gabor滤波中）的差异，可解释为构图方向不同，而非技法差异。本数据集为创作过程的一致性提供了强有力的证据，支持其可能出自梵高晚期之手的假设。

---

📖 引用

如使用本数据集，请引用：

孟武思 (Haruthai). 2025. TreeRoots1890_vs_TreeOilPainting_18Tech_Xray_Ai_Analysis. Hugging Face.