HaruthaiAi/Zhang_Xian_TorqueDataset_README

🎨 TorqueBrush Forensics – Zhang Xian Portrait (1928)

🧭 Overview

This dataset presents high-resolution visual and computational analyses of the oil portrait attributed to Zhang Xian (1898–1936).

It uses TorqueBrush, an AI-assisted method developed to analyze torque, stroke isolation, and brush dynamics. The 1928 portrait reveals exceptional anatomical precision and torque patterns consistent with advanced draftsmanship.

📌 Now part of global forensic research: 🔗 https://huggingface.co/datasets/HaruthaiAi/zhangxian_1928_portrait_discovery_storm_society_modern_art_pioneer

📁 Files Included

zhangxian_original.png – Grayscale version of the painting

torque_analysis_grid.png – 6-panel composite showing:

Original Image

Edge Magnitude

Direction Field

Torque Field

Stroke Isolation

Acceleration Map

🎯 Purpose

1. Scientific Preservation – Torque signature creation

2. Cultural Revival – Support Chinese art heritage

3. Collaborative Verification – Enable AI + Humanities synergy

🧪 How to Run TorqueBrush

Step 1: Install Required Libraries

pip install opencv-python numpy matplotlib scipy

Step 2–6: Python Code

import cv2
import numpy as np
import matplotlib.pyplot as plt
from scipy.ndimage import sobel, gaussian_filter
from google.colab import files

uploaded = files.upload()
image_path = list(uploaded.keys())[0]
img = cv2.imread(image_path, cv2.IMREAD_GRAYSCALE)
img = cv2.resize(img, (800, int(img.shape[0] * 800 / img.shape[1])))

sobel_x = sobel(img, axis=0)
sobel_y = sobel(img, axis=1)
edge_magnitude = np.hypot(sobel_x, sobel_y)
direction_field = np.arctan2(sobel_y, sobel_x)
torque_field = sobel_x * sobel_y
blurred = gaussian_filter(img, sigma=1)
laplacian = cv2.Laplacian(blurred, cv2.CV_64F)
binary = cv2.threshold(blurred, 128, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]

plt.figure(figsize=(20, 10))
plt.subplot(2, 3, 1); plt.imshow(img, cmap='gray'); plt.title("Original Image")
plt.subplot(2, 3, 2); plt.imshow(edge_magnitude, cmap='inferno'); plt.title("Edge Magnitude")
plt.subplot(2, 3, 3); plt.imshow(direction_field, cmap='twilight'); plt.title("Direction Field")
plt.subplot(2, 3, 4); plt.imshow(torque_field, cmap='coolwarm'); plt.title("Torque Field")
plt.subplot(2, 3, 5); plt.imshow(binary, cmap='gray'); plt.title("Stroke Isolation")
plt.subplot(2, 3, 6); plt.imshow(laplacian, cmap='magma'); plt.title("Acceleration Map")
plt.tight_layout()
plt.show()

🖼️ Sample Output

📊 TorqueBrush Grid – Zhang Xian Portrait

📝 Curator’s Note (Haruthai Muangbunsri)

This portrait was the first key to unlocking the Tree Oil Painting — a discovery that connects Eastern expressionism with Western post-impressionist energy through torque.

🔖 Tags

zhangxian torquebrush art_ai china_republican_era historical_portrait digital_humanities

---

License: CC-BY 4.0 Created: May 23, 2025 Maintained by: Haruthai Muangbunsri (Thailand)