Design and Evaluation of ADANet: A High-Fidelity Motion Acquisition Framework for Assistive Gesture-Based Interfaces

Accurate and low-latency gesture recognition is critical for real-time assistive technologies that enable individuals with motor impairments to interact more intuitively with their environment. However, current systems often suffer from poor signal fidelity, limited adaptability, and high computatio...

Full description

Saved in:
Bibliographic Details
Main Authors: Md Ettashamul Haque, Atique Tajwar, Akm Azad, Salem A. Alyami, Md Mehedi Hasan
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11091288/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849245971286851584
author Md Ettashamul Haque
Atique Tajwar
Akm Azad
Salem A. Alyami
Md Mehedi Hasan
author_facet Md Ettashamul Haque
Atique Tajwar
Akm Azad
Salem A. Alyami
Md Mehedi Hasan
author_sort Md Ettashamul Haque
collection DOAJ
description Accurate and low-latency gesture recognition is critical for real-time assistive technologies that enable individuals with motor impairments to interact more intuitively with their environment. However, current systems often suffer from poor signal fidelity, limited adaptability, and high computational overhead. To address these limitations, this article presents ADANet &#x2014; Advanced Disability Assistive Neural Network-driven framework for accelerometer-based gesture recognition, tailored for wearable assistive applications. The proposed system combines a high-fidelity ADXL335 triaxial accelerometer with an ESP32 microcontroller, forming a lightweight and cost-efficient motion acquisition pipeline. A structured preprocessing architecture is developed, incorporating zero-lag Butterworth filtering, entropy-based temporal smoothing, and computation of 33 handcrafted statistical features, including axis-specific jerk, signal magnitude area (SMA), and range-normalized entropy. ADANet&#x2019;s compact yet expressive neural architecture is optimized through comprehensive ablation studies to ensure strong generalization with minimal latency. Our model was trained on data from 21 healthy participants(12 male &#x0026; 9 female) performing eight functional gestures demonstrate a test accuracy of 84.47%, F1-score of 0.84, and AUC<inline-formula> <tex-math notation="LaTeX">$\geq 0.96$ </tex-math></inline-formula>, outperforming traditional classifiers such as XGBoost, SVM, and KNN across multiple evaluation metrics. This work validates a robust end-to-end instrumentation-to-inference framework and establishes ADANet as a practical solution for embedded gesture recognition in assistive devices, mostly for disabilities, where signal integrity, adaptability, and inference efficiency are jointly critical.
format Article
id doaj-art-87c456d0aa064d44a7486ca499b2db43
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-87c456d0aa064d44a7486ca499b2db432025-08-20T03:58:39ZengIEEEIEEE Access2169-35362025-01-011313303513305210.1109/ACCESS.2025.359207911091288Design and Evaluation of ADANet: A High-Fidelity Motion Acquisition Framework for Assistive Gesture-Based InterfacesMd Ettashamul Haque0Atique Tajwar1Akm Azad2https://orcid.org/0000-0002-5251-2214Salem A. Alyami3https://orcid.org/0000-0002-5507-9399Md Mehedi Hasan4https://orcid.org/0000-0003-4326-4332Department of Robotics and Mechatronics Engineering, University of Dhaka, Dhaka, BangladeshDepartment of Robotics and Mechatronics Engineering, University of Dhaka, Dhaka, BangladeshDepartment of Mathematics and Statistics, Faculty of Science, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh, Saudi ArabiaDepartment of Mathematics and Statistics, Faculty of Science, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh, Saudi ArabiaDepartment of Robotics and Mechatronics Engineering, University of Dhaka, Dhaka, BangladeshAccurate and low-latency gesture recognition is critical for real-time assistive technologies that enable individuals with motor impairments to interact more intuitively with their environment. However, current systems often suffer from poor signal fidelity, limited adaptability, and high computational overhead. To address these limitations, this article presents ADANet &#x2014; Advanced Disability Assistive Neural Network-driven framework for accelerometer-based gesture recognition, tailored for wearable assistive applications. The proposed system combines a high-fidelity ADXL335 triaxial accelerometer with an ESP32 microcontroller, forming a lightweight and cost-efficient motion acquisition pipeline. A structured preprocessing architecture is developed, incorporating zero-lag Butterworth filtering, entropy-based temporal smoothing, and computation of 33 handcrafted statistical features, including axis-specific jerk, signal magnitude area (SMA), and range-normalized entropy. ADANet&#x2019;s compact yet expressive neural architecture is optimized through comprehensive ablation studies to ensure strong generalization with minimal latency. Our model was trained on data from 21 healthy participants(12 male &#x0026; 9 female) performing eight functional gestures demonstrate a test accuracy of 84.47%, F1-score of 0.84, and AUC<inline-formula> <tex-math notation="LaTeX">$\geq 0.96$ </tex-math></inline-formula>, outperforming traditional classifiers such as XGBoost, SVM, and KNN across multiple evaluation metrics. This work validates a robust end-to-end instrumentation-to-inference framework and establishes ADANet as a practical solution for embedded gesture recognition in assistive devices, mostly for disabilities, where signal integrity, adaptability, and inference efficiency are jointly critical.https://ieeexplore.ieee.org/document/11091288/Gesture recognitionaccelerometersignal acquisition systemneural networkmachine learninghuman–computer interface (HCI)
spellingShingle Md Ettashamul Haque
Atique Tajwar
Akm Azad
Salem A. Alyami
Md Mehedi Hasan
Design and Evaluation of ADANet: A High-Fidelity Motion Acquisition Framework for Assistive Gesture-Based Interfaces
IEEE Access
Gesture recognition
accelerometer
signal acquisition system
neural network
machine learning
human–computer interface (HCI)
title Design and Evaluation of ADANet: A High-Fidelity Motion Acquisition Framework for Assistive Gesture-Based Interfaces
title_full Design and Evaluation of ADANet: A High-Fidelity Motion Acquisition Framework for Assistive Gesture-Based Interfaces
title_fullStr Design and Evaluation of ADANet: A High-Fidelity Motion Acquisition Framework for Assistive Gesture-Based Interfaces
title_full_unstemmed Design and Evaluation of ADANet: A High-Fidelity Motion Acquisition Framework for Assistive Gesture-Based Interfaces
title_short Design and Evaluation of ADANet: A High-Fidelity Motion Acquisition Framework for Assistive Gesture-Based Interfaces
title_sort design and evaluation of adanet a high fidelity motion acquisition framework for assistive gesture based interfaces
topic Gesture recognition
accelerometer
signal acquisition system
neural network
machine learning
human–computer interface (HCI)
url https://ieeexplore.ieee.org/document/11091288/
work_keys_str_mv AT mdettashamulhaque designandevaluationofadanetahighfidelitymotionacquisitionframeworkforassistivegesturebasedinterfaces
AT atiquetajwar designandevaluationofadanetahighfidelitymotionacquisitionframeworkforassistivegesturebasedinterfaces
AT akmazad designandevaluationofadanetahighfidelitymotionacquisitionframeworkforassistivegesturebasedinterfaces
AT salemaalyami designandevaluationofadanetahighfidelitymotionacquisitionframeworkforassistivegesturebasedinterfaces
AT mdmehedihasan designandevaluationofadanetahighfidelitymotionacquisitionframeworkforassistivegesturebasedinterfaces