BdSLW60: A word-level bangla sign language dataset

Multimedia Tools and Applications, Год журнала: 2025, Номер unknown

Опубликована: Апрель 14, 2025

Язык: Английский

---

Convolutional Block Attention Augmented Convolutional Neural Network for Indian Sign Language Recognition System

Smart innovation, systems and technologies, Год журнала: 2025, Номер unknown, С. 215 - 226

Опубликована: Янв. 1, 2025

Язык: Английский

---

Bangla Sign Language Recognition With Multimodal Deep Learning Fusion

Engineering Reports, Год журнала: 2025, Номер 7(4)

Опубликована: Апрель 1, 2025

ABSTRACT Sign languages are conveyed through hand gestures and body language; different cultures have unique sign language representations. It is difficult for the general population to interpret all these variations. A Bangla recognition system can help mitigate this problem. In Bangladesh, approximately 13.7 million people affected by hearing impairments, highlighting importance of enhancing system. study, a merged dataset consists 200‐word classes 7,000 videos performed 16 signers. aspect that it includes two modalities. Initially, we extracted frames audio from videos. We then employed frameworks capture keypoints frames. Additionally, contributed expanding creating custom adds more samples per class. During preprocessing, augmented training data, improving our model's learning capacity reducing overfitting. developed leveraging various deep techniques, simultaneously working with proposed CNN‐LSTM model pose estimation integrated VGGish 2D CNN techniques process in multimodal model. The attained an accuracy 87.08% using OpenPose framework. applied ViT achieved best performance 88.52% 88.39% F1 score predict directly. technique delivered result dataset, achieving 81.91% 80.42% score. Finally, late fusion approach combining networks employing data performance, 94.71% 94.52%

Язык: Английский

---

Interpretation of Indian Sign Language to Text and Speech to Communicate with Speech and Hearing Impaired Community

Procedia Computer Science, Год журнала: 2025, Номер 258, С. 1980 - 1992

Опубликована: Янв. 1, 2025

Язык: Английский

---

Efficient Models Based on Deep Learning Technique for Indian Sign Language

Procedia Computer Science, Год журнала: 2025, Номер 258, С. 318 - 331

Опубликована: Янв. 1, 2025

Язык: Английский

---

Pakistan Sign Language Recognition: From Videos to Images

Signal Image and Video Processing, Год журнала: 2025, Номер 19(8)

Опубликована: Июнь 2, 2025

Язык: Английский

---

Machine Learning-based Intrusion Detection System Through WPA3 Protocol in Smart Contract System

International Journal of Innovative Science and Research Technology (IJISRT), Год журнала: 2024, Номер unknown, С. 2926 - 2942

Опубликована: Апрель 17, 2024

Nowadays, the Internet has become one of basic human needs professionals. With massive number devices, reliability, and security will be crucial in coming ages. Routers are common to provide us with internet. These routers can operated different modes. Some use Wifi Security Protocol (WPA) or WPA2, Alliance introduced WPA3 on 25 June 2018. There a lot papers regarding Smart Contract (SC)–based IDS as well Machine Learning-based IDS. Very few discuss combining SC ML-based for authentication processes. In this paper, we how ML plays vital role authentication. Also, play embedded system so that existing vulnerabilities WPA2 reduced 99.62%.

Язык: Английский

---

Computer Vision Based Bangla Sign Language Recognition Using Transfer Learning

Опубликована: Май 17, 2024

Язык: Английский

---

Real-Time Sign Language Detection: Empowering the Disabled Community

MethodsX, Год журнала: 2024, Номер 13, С. 102901 - 102901

Опубликована: Авг. 8, 2024

Interaction and communication for normal human beings are easier than a person with disabilities like speaking hearing who may face problems other people. Sign Language helps reduce this gap between disabled person. The prior solutions proposed using several deep learning techniques, such as Convolutional Neural Networks, Support Vector Machines, K-Nearest Neighbors, have either demonstrated low accuracy or not been implemented real-time working systems. This system addresses both issues effectively. work extends the difficulties faced while classifying characters in Indian Language(ISL). It can identify total of 23 hand poses ISL. uses pre-trained VGG16 Convolution Network(CNN) an attention mechanism. model's training Adam optimizer cross-entropy loss function. results demonstrate effectiveness transfer ISL classification, achieving 97.5 % 99.8 plus mechanism.•Enabling quick accurate sign language recognition help trained model mechanism.•The does require any external gloves sensors, which to eliminate need physical sensors simplifying process reduced costs.•Real-time processing makes more helpful people disabilities, making it them communicate humans.

Язык: Английский

---

Unveiling the Power of Machine Learning and Deep Learning in Advancing American Sign Language Recognition

Опубликована: Апрель 17, 2024

American Sign Language (ASL) serves as a crucial visual-gestural language for the Deaf community, characterized by its intricate syntax expressed through gestures, facial expressions, and body language. Despite complexity, ASL presents challenges in realm of deep learning machine due to dynamic nature, varied signing methods, absence standardized datasets. The scarcity labeled data further complicates recognition, hindering capture subtle nuances such hand movements expressions. This research study intends address these developing models that convert images into text, thereby advancing field sign recognition. In this study, efficacy OpenCV's image classification techniques, namely K-Nearest Neighbors, Naive Bayes, Support Vector Machines, Logistic Regression, is demonstrated ASL-related tasks. Remarkably, achieve accuracies ranging from 75% an impressive 80%. second experiment focuses on ASL-specific applications, leveraging superior performance Machines (SVM). Utilizing dataset encompassing four alphabets (A-D), nine words, ten integers, combined approach SVM YOLOv5 achieves object detection accuracy 81.16%, surpassing MediaPipe. Furthermore, delves extension strategies, revealing noteworthy 92.11% attained when combining with Convolutional Neural Networks (CNN). Additionally, 98.01% achieved utilizing conjunction Long Short-Term Memory (LSTM). comprehensive investigation not only contributes valuable insights development effective communication systems but also significantly expands our understanding learning, identification.

Язык: Английский

---