Related Experiment Video
Updated: Jun 12, 2026
08:15
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
Published on: March 28, 2025
Effective optical processor for computing image moments at TV rate: use in handwriting recognition.
Applied Optics
|May 22, 2010
Summary
This study introduces an optical processor for real-time image moment computation. The system effectively generates moments for pattern recognition applications, demonstrating practical utility.
Area of Science:
- Optics
- Computer Vision
- Image Processing
Background:
- Traditional image analysis methods can be computationally intensive.
- Real-time processing is crucial for dynamic pattern recognition tasks.
Purpose of the Study:
- To describe an incoherent multiplexed optical processor for real-time image moment computation.
- To evaluate the system's accuracy and practical application in pattern recognition.
Main Methods:
- Utilized an incoherent multiplexed optical processor to compute image moments in real time.
- Employed a TV camera for image sensing and a photodetector array for moment detection.
- Integrated a microcomputer for data collection and subsequent pattern recognition analysis.
Main Results:
- Successfully generated the first ten image moments in parallel in real time.
- Achieved an effective system accuracy of 2%, considering calibration and resolution limitations.
- Demonstrated practical application by processing one hundred handwritten pages in real time.
Conclusions:
- The developed optical processor is a practical and effective tool for generating image moments.
- The system shows potential for real-time pattern recognition tasks, particularly with handwritten documents.
