Ensemble Learning of Multiple Deep CNNs Using Accuracy-Based Weighted Voting for ASL Recognition
More than four million people worldwide suffer from hearing loss. Recently, new CNNs and deep ensemble-learning technologies have brought promising opportunities to the image-recognition field, so many studies aiming to recognize American Sign Language (ASL) have been conducted to help these people...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/22/11766 |
| _version_ | 1797465901588545536 |
|---|---|
| author | Ying Ma Tianpei Xu Seokbung Han Kangchul Kim |
| author_facet | Ying Ma Tianpei Xu Seokbung Han Kangchul Kim |
| author_sort | Ying Ma |
| collection | DOAJ |
| description | More than four million people worldwide suffer from hearing loss. Recently, new CNNs and deep ensemble-learning technologies have brought promising opportunities to the image-recognition field, so many studies aiming to recognize American Sign Language (ASL) have been conducted to help these people express their thoughts. This paper proposes an ASL Recognition System using Multiple deep CNNs and accuracy-based weighted voting (ARS-MA) composed of three parts: data preprocessing, feature extraction, and classification. Ensemble learning using multiple deep CNNs based on LeNet, AlexNet, VGGNet, GoogleNet, and ResNet were set up for the feature extraction and their results were used to create three new datasets for classification. The proposed accuracy-based weighted voting (AWV) algorithm and four existing machine algorithms were compared for the classification. Two parameters, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>λ</mi></semantics></math></inline-formula>, are introduced to increase the accuracy and reduce the testing time in AWV. The experimental results show that the proposed ARS-MA achieved 98.83% and 98.79% accuracy on the ASL Alphabet and ASLA datasets, respectively. |
| first_indexed | 2024-03-09T18:29:03Z |
| format | Article |
| id | doaj.art-22ef7015d14e40f19e76454b814a287e |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T18:29:03Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-22ef7015d14e40f19e76454b814a287e2023-11-24T07:40:34ZengMDPI AGApplied Sciences2076-34172022-11-0112221176610.3390/app122211766Ensemble Learning of Multiple Deep CNNs Using Accuracy-Based Weighted Voting for ASL RecognitionYing Ma0Tianpei Xu1Seokbung Han2Kangchul Kim3Department of Computer Science and Engineering, Chonnam National University, Yeosu 59626, Republic of KoreaDepartment of Computer Science and Engineering, Chonnam National University, Yeosu 59626, Republic of KoreaDepartment of Electronic Engineering, Gyeongsang National University, Jinju 52828, Republic of KoreaDepartment of Computer Science and Engineering, Chonnam National University, Yeosu 59626, Republic of KoreaMore than four million people worldwide suffer from hearing loss. Recently, new CNNs and deep ensemble-learning technologies have brought promising opportunities to the image-recognition field, so many studies aiming to recognize American Sign Language (ASL) have been conducted to help these people express their thoughts. This paper proposes an ASL Recognition System using Multiple deep CNNs and accuracy-based weighted voting (ARS-MA) composed of three parts: data preprocessing, feature extraction, and classification. Ensemble learning using multiple deep CNNs based on LeNet, AlexNet, VGGNet, GoogleNet, and ResNet were set up for the feature extraction and their results were used to create three new datasets for classification. The proposed accuracy-based weighted voting (AWV) algorithm and four existing machine algorithms were compared for the classification. Two parameters, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>λ</mi></semantics></math></inline-formula>, are introduced to increase the accuracy and reduce the testing time in AWV. The experimental results show that the proposed ARS-MA achieved 98.83% and 98.79% accuracy on the ASL Alphabet and ASLA datasets, respectively.https://www.mdpi.com/2076-3417/12/22/11766ASLweighted votingCNNensemble learning |
| spellingShingle | Ying Ma Tianpei Xu Seokbung Han Kangchul Kim Ensemble Learning of Multiple Deep CNNs Using Accuracy-Based Weighted Voting for ASL Recognition Applied Sciences ASL weighted voting CNN ensemble learning |
| title | Ensemble Learning of Multiple Deep CNNs Using Accuracy-Based Weighted Voting for ASL Recognition |
| title_full | Ensemble Learning of Multiple Deep CNNs Using Accuracy-Based Weighted Voting for ASL Recognition |
| title_fullStr | Ensemble Learning of Multiple Deep CNNs Using Accuracy-Based Weighted Voting for ASL Recognition |
| title_full_unstemmed | Ensemble Learning of Multiple Deep CNNs Using Accuracy-Based Weighted Voting for ASL Recognition |
| title_short | Ensemble Learning of Multiple Deep CNNs Using Accuracy-Based Weighted Voting for ASL Recognition |
| title_sort | ensemble learning of multiple deep cnns using accuracy based weighted voting for asl recognition |
| topic | ASL weighted voting CNN ensemble learning |
| url | https://www.mdpi.com/2076-3417/12/22/11766 |
| work_keys_str_mv | AT yingma ensemblelearningofmultipledeepcnnsusingaccuracybasedweightedvotingforaslrecognition AT tianpeixu ensemblelearningofmultipledeepcnnsusingaccuracybasedweightedvotingforaslrecognition AT seokbunghan ensemblelearningofmultipledeepcnnsusingaccuracybasedweightedvotingforaslrecognition AT kangchulkim ensemblelearningofmultipledeepcnnsusingaccuracybasedweightedvotingforaslrecognition |