Kratos-AI/Handwritten-Physics-Notes-Dataset

English Handwritten Physics Notes Dataset

This dataset contains high-resolution images of handwritten physics notes written in English. The collection includes theoretical explanations, formulas, diagrams, derivations, and problem-solving steps. It is designed to support AI research in handwriting recognition, scientific OCR, and document understanding for physics and STEM education.

Contact

For queries or collaborations related to this dataset, contact:

• [email protected]
• [email protected]

Supported Tasks

• Task Categories:

  • Image Classification
  • Text and Equation Recognition (OCR)
  • Diagram and Symbol Detection
  • Document Layout Understanding

• Supported Tasks:

  • Recognition of handwritten English physics notes and equations
  • Classification of diagrams and visual representations (e.g., force diagrams, circuits)
  • OCR for formulas, units, and explanatory text
  • Structural analysis of mixed-text and equation layouts
  • Research in AI-powered physics education and content digitization

Languages

• Primary Language: English
• Secondary Presence: Mathematical and physical symbols (e.g., F=ma, ∆E, v, λ, etc.)

Dataset Creation

Curation Rationale

The dataset was curated to enable the development of AI systems capable of understanding complex handwritten physics content combining text, formulas, and diagrams. It facilitates progress in educational technology, note digitization, and automated content comprehension in science education.

Source Data

• Contributors: Students, educators, and volunteers who shared handwritten class or study notes
• Collection Process: Notes were scanned or photographed. All personal identifiers were removed prior to inclusion.

Other Known Limitations

• Bias: Overrepresentation of student-level and academic notes
• Image Quality: Variations due to lighting, ink, and paper texture
• Content Scope: Focused mainly on classical and high school physics; advanced topics may be limited

Intended Uses

✅ Direct Use

• Training AI models for handwriting and equation recognition in physics
• Academic research in scientific OCR and content understanding
• Digitization of educational physics materials
• Development of AI tutoring or smart note apps for STEM learners

❌ Out-of-Scope Use

• Identifying contributors based on handwriting or personal patterns
• Commercial reuse of handwriting without permission
• Use in surveillance or behavioral analysis contexts

License

CC BY 4.0