A comprehensive study on sign languages recognition systems using (SVM, KNN, CNN and ANN)

Agha, Rawan A. Al Rashid; Sefer, Muhammed N.; Fattah, Polla

doi:10.1145/3279996.3280024

Cited by 9 publications

(6 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Te total dataset contains 1.25 million samples [75] 2018 Indian Sign Language (ISL) Digits 0 to 9 and alphabets a to z were taken for the experiment [79] 2018 Indian Sign Language (ISL) Digits 0 to 9 and alphabets a to z were taken for the experiment [83] 2018 Custom built. Indian Sign Language 18 signs with each sign by 10 diferent signers recorded [71] 2018 Indian Sign Language American Sign Language British Sign Language Turkish Sign Language (i) ISL dataset: used SVM for this dataset Contains 4 signs, that is, A, B, C, and the word "Hello" (ii) ASL dataset: used KNN for this dataset Contains 10 ASL fngerspelling alphabets from a to i and k.…”

Section: Rq7: Which Models Have Demonstrated Bettermentioning

confidence: 99%

See 1 more Smart Citation

Exploring Sign Language Detection on Smartphones: A Systematic Review of Machine and Deep Learning Approaches

Alam,

Hameed,

Ziar

2024

Advances in Human-Computer Interaction

View full text Add to dashboard Cite

In this modern era of technology, most of the accessibility issues are handled with the help of smart devices and cutting-edge gadgets. Smartphones play a crucial role in addressing various accessibility challenges, including voice recognition, sign language detection and interpretation, navigation systems, speech-to-text conversion, and vice versa, among others. They are computationally powerful enough to handle and run numerous machine and deep learning applications. Among various accessibility challenges, speech disorders represent a disability where individuals struggle to communicate verbally. Similarly, hearing loss is a disability that impairs an individual’s ability to hear, necessitating reliance on gestures for communication. A significant challenge encountered by people with speech disorders, hearing loss, or both is their inability to effectively convey or receive messages from others. Hence, these individuals heavily depend on the sign language (a gesture-based communication) method, typically involving hand movements and expressions. To the best of our knowledge, there are currently no comprehensive review and/or survey articles available that cover the literature on speech disabilities and sign language detection and interpretation via smartphones utilizing machine learning and/or deep learning approaches. This study fills the gap in the literature by analyzing research publications on speech disabilities, published from 2012 to July 2023. A rigorous search and standard strategy for formulating the literature along with a well-defined theoretical framework for results and findings have been used. The paper has implications for practitioners and researchers working in accessibilities in general and smart/intelligent gadgets and applications for speech-disabled people in specific.

show abstract

Section: Rq7: Which Models Have Demonstrated Bettermentioning

confidence: 99%

“…Sign languages targeted 40,44,45,50,57,65,68,71,78,81,84,86,88,91,96,98,100,105,106,110,112,[117][118][119][120]132]. …”

mentioning

confidence: 99%

Exploring Sign Language Detection on Smartphones: A Systematic Review of Machine and Deep Learning Approaches

Alam,

Hameed,

Ziar

2024

Advances in Human-Computer Interaction

View full text Add to dashboard Cite

show abstract

“…Various feature extraction methods and insight to the image‐based SLR system are described by Bhavsar 27 . A comprehensive approach is provided by Al Rashid Agha et al 28 on four different approaches namely support vector machine (SVM), k‐nearest neighbor (KNN), convolutional neural network (CNN) and artificial neural network (ANN) for implementing sign language recognition. He presented deep learning‐based approaches for action and gesture recognition of image sequence; he also reviews the details of main datasets, fusion streams, and different proposed architectures.…”

Section: Motivations and Problem Statementmentioning

confidence: 99%

A review on computational methods based automated sign language recognition system for hearing and speech impaired community

Robert

Hemanth

2023

Concurrency and Computation

View full text Add to dashboard Cite

Summary The recent advancements in computer vision and deep learning have led to promising progress in various motion detection and gesture recognition methods. Thriving efforts in the field of sign language recognition (SLR) during recent years led to interaction between humans and computer systems. Contributing a real‐time automated Sign Language recognition system will be a remarkable treasure for hearing and speech‐impaired people that will break the barriers of interaction with the real world. Albeit several research works are accomplished in sign language recognition, there is still a demand for developing a real‐time automated sign language recognition system. Compared to other methodologies, the techniques adopted may have advantages and disadvantages, and they vary from researcher to researcher. There are still some issues with employing these SLR models and procedures regularly, even though numerous research studies have been conducted to determine the most acceptable methods and models for sign language recognition. It gets more expensive and difficult in terms of resources when the developed automated SLR system becomes available as a product. Hence, for the welfare of the hearing and speech‐impaired community, the researchers are still endeavoring way to find a cost‐effective method. This work brings forth the challenges faced by scientists in developing a cost‐effective commercial prototype for the hearing and speech‐impaired community. This paper also explores and analyses various deep learning techniques and methods used in developing a sign language recognition system. The objective behind this work is to identify the best method which produces high accuracy in developing a cost‐effective sign language recognition system aiding communication between signers and non‐signers so that they can be a part of growing technology.

show abstract

“…Convolutional Neural Network (CNN), which is a mult ilayered feed-forward artificial neural network, is one of the most frequently used methods in image processing [17]. CNN, which has an important place in the field of deep learning, also shows success in various applications such as face recognition [18], object classification [19], speech detection [20], and sign language recognition [21,22]. The most preferred CNN for dealing with computer vision problems is architecturally formed by adding convolution, pooling (sampling), and fully connected layers to the basic layers [23,24].…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

Motion Control of the Robot Arm Manufactured with a Three-Dimensional Printer and Hardness Detection of Objects

Aksoy

Özsoy

YÜCEL>

et al. 2022

Bilişim Teknolojileri Dergisi

View full text Add to dashboard Cite

Çalışmada 3D baskı teknolojilerinden Fused Deposition Modeling (FDM) yazıcı kullanılarak robotik kol üretilmiştir. Üretilen robot kolun görüntü işleme teknikleri ve makine öğrenme algoritmaları kullanarak dokunsal algılama ve hareket planlaması araştırılmıştır. Bu çalışmanın amacı, robotik kolun kontrolsüz kuvvet uygulamasını engellemek ve dokunsal kavrama sorunlarını çözmek için görüntü işleme teknikleri ve derin öğrenme algoritmaları kullanılarak yenilikçi yaklaşımların araştırılması ve uygulanmasıdır. Bu çalışmada, CAD programı ile tasarımı gerçekleştirilmiş parçaların FDM tipi üç boyutlu yazıcı kullanılarak katı modelleri alınmış ve montaj için uygun hale getirilmiştir. Montajı tamamlanan robotik elin kontrol sistemi ise temel olarak Raspberry Pi kontrol kartı, servo motorlar, basınç sesörleri ve kameradan oluşmaktadır. Robotik kola ait her parmak ucuna yerleştirilen basınç sensörleri ile ürünün sertliği ölçülerek dokunsal algılama işlemi gerçekleştirilmiştir. Raspberrry pi kontrol kartı kullanılarak sensörlerden alınan veriler işlenmekte ve servo motorlara uygun hareket ve kavrama basınç bilgisi gönderilmektedir. Kamera kullanılarak elde edilen insan elinin olası hareketleri ile robotik kol için referans bir veri seti hazırlanmıştır. Veri setine ait görüntüler üzerinde Gaussian filtreleme yöntemi kullanılarak görüntü işleme sağlanmıştır. Bununla birlikte veri seti üzerinde makine öğrenme algoritmaları kullanarak robotik kolun hareket açısal konumu optimize edilmiş ve HitNet, CNN, Kapsül Ağları ve Naive Bayes derin öğrenme modelleri kullanılarak robot kolun hareket planlanması %90 doğruluk oranı ile sınıflandırılmıştır. Performans değerlendirme kriterlerine göre başarıları kıyaslanan derin öğrenme modelleri arasında, robotik kolun hareket planlaması için; HitNET algoritması ile 97.23%, CNN ile 97.48%, Capsnet algoritması ile %98,58 ve Naive Bayes modeli ile %98.61 doğruluk oranı elde edilmiştir. Performans değerlendirme kriterleri sonucunda; Naive Bayes modelinin %98.61 doğruluk, %98.63 özgüllük, %98.65 duyarlılık, 1.39 hata oranı ve %68.64 F-ölçüsü değeri ile diğer modellere göre daha başarılı sonuç verdiği gözlemlenmiştir.

show abstract

A comprehensive study on sign languages recognition systems using (SVM, KNN, CNN and ANN)

Cited by 9 publications

References 13 publications

Exploring Sign Language Detection on Smartphones: A Systematic Review of Machine and Deep Learning Approaches

Exploring Sign Language Detection on Smartphones: A Systematic Review of Machine and Deep Learning Approaches

A review on computational methods based automated sign language recognition system for hearing and speech impaired community

Motion Control of the Robot Arm Manufactured with a Three-Dimensional Printer and Hardness Detection of Objects

Contact Info

Product

Resources

About