Enhancement Of You Only Look Once v6 (YOLOv6) Algorithm Applied to Sign Language Recognition System

Louisse Andrea Mae Macugay Toribio; Kennett Lim Miralles; Dan Michael A. Cortez; Khatalyn E. Mata

doi:10.36948/ijfmr.2026.v08i02.72444

Enhancement Of You Only Look Once v6 (YOLOv6) Algorithm Applied to Sign Language Recognition System

Author(s)	Ms. Louisse Andrea Mae Macugay Toribio, Mr. Kennett Lim Miralles, Dr. Dan Michael A. Cortez, Dr. Khatalyn E. Mata
Country	Philippines
Abstract	This study enhances the You Only Look Once version 6 (YOLOv6) algorithm, specifically optimized for an American Sign Language (ASL) recognition system to bridge communication gaps for the Deaf community. Current YOLO-based real-time object detection models often struggle with image distortion from fixed-scale resizing, poor localization of small objects using standard Intersection over Union (IoU), and inaccurate detection of overlapping gestures due to standard Non-Maximum Suppression (NMS). This research addresses these limitations by integrating an Adaptive Image Resizing technique to preserve aspect ratios, implementing Distance-IoU (DIoU) to improve bounding box alignment for small objects, and utilizing Soft Non-Maximum Suppression (Soft-NMS) to retain valid detections in overlapping scenarios. Experimental results after simulations demonstrate that the enhanced algorithm (e-YOLOv6) significantly outperforms existing models across all key metrics. The e-YOLOv6 achieved a mean Average Precision (mAP50) of 83.6%, a marked increase over the 54.4% and 71.1% scores of the baseline YOLOv6 versions. Furthermore, precision reached 98.3%, and the F1-score improved to 81.1%, representing a 13.3% gain over the highest-performing baseline model. These results confirm that preventing image distortion and utilizing distance-based penalty functions are critical for high-accuracy gesture recognition. With this, e-YOLOv6 provides a superior framework for real-time sign language interpretation by successfully mitigating the "miss rate" for occluded and distorted gestures, facilitating more inclusive and independent communication for Deaf individuals.
Keywords	YOLOv6, Sign Language Recognition, Adaptive Image Resizing, Distance-IoU (DIoU), Soft-NMS, American Sign Language (ASL), Computer Vision
Field	Computer > Data / Information
Published In	Volume 8, Issue 2, March-April 2026
Published On	2026-04-09
DOI	https://doi.org/10.36948/ijfmr.2026.v08i02.72444

View / Download PDF File

E-ISSN 2582-2160

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJFMR DOI prefix is
10.36948/ijfmr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 8 Isu 2 Cover Page Vol 8 Isu 1 Cover Page Vol 7 Isu 6

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Enhancement Of You Only Look Once v6 (YOLOv6) Algorithm Applied to Sign Language Recognition System

Share this