# Indian Sign Language Recognition Using Eigen Value Weighted Euclidean Distance Based Classification Technique

**Authors:** Joyeeta Singha, Karen Das  
**arXiv:** [1303.0634](https://arxiv.org/abs/1303.0634)  
**Date:** 2013-03-04  
**Summary page:** https://www.summarizepaper.com/en/paper/1303.0634v1/

## Key points

- Significant growth in research and development of Sign Language Recognition techniques
- Proposed system utilizes Eigen value weighted Euclidean distance as a classification technique
- System consists of four main parts: Skin Filtering, Hand Cropping, Feature Extraction, and Classification
- System accurately identifies and classifies different signs in Indian Sign Language
- Evaluation of system using twenty-four signs with ten samples each resulted in a recognition rate of 97 percent
- Novel approach leveraging Eigen value weighted Euclidean distance for classification
- High recognition rate demonstrates potential for practical applications in real-world scenarios involving Indian Sign Language
- Paper highlights ongoing progress in Sign Language Recognition and presents a promising system for accurately recognizing various signs in Indian Sign Language.

## Summary

In the field of Sign Language Recognition, there has been significant growth in research and the development of new techniques. In a recent paper titled "Indian Sign Language Recognition Using Eigen Value Weighted Euclidean Distance Based Classification Technique," authors Joyeeta Singha and Karen Das propose a system that utilizes Eigen value weighted Euclidean distance as a classification technique for recognizing various Sign Languages of India. The proposed system consists of four main parts: Skin Filtering, Hand Cropping, Feature Extraction, and Classification. These components work together to accurately identify and classify different signs in Indian Sign Language. To evaluate the effectiveness of their system, the authors considered twenty-four signs with ten samples each; resulting in two hundred forty images being used for recognition. The obtained recognition rate was an impressive 97 percent. This research contributes to the advancement of Sign Language Recognition by introducing a novel approach that leverages Eigen value weighted Euclidean distance as a classification technique. The system's high recognition rate demonstrates its potential for practical applications in real-world scenarios involving Indian Sign Language. Overall, this paper highlights the ongoing progress in Sign Language Recognition and presents a promising system that shows great potential for accurately recognizing various signs in Indian Sign Language.