Cited by CNN Algorithm with SIFT to Enhance the Arabic Sign Language Recognition

CNN Algorithm with SIFT to Enhance the Arabic Sign Language Recognition DOI

et al.

International Journal of Emerging Science and Engineering, Journal Year: 2024, Volume and Issue: 12(10), P. 12 - 17

Published: Sept. 24, 2024

Sign language is used as a primary means of communication by millions people who suffer from hearing problems. The unhearing visual to interact with each other, Represented in sign language. There are features that the impaired use understand which difficult for normal understand. Therefore, deaf will struggle society. This research aims introduce system recognizing hand gestures Arabic Language (ArSL) through training Convolutional Neural Network (CNN) on images ArSL launched University Prince Mohammad Bin Fahd, Saudi Arabia. A Scale Invariant Feature Transform (SIFT) algorithm creating feature vectors contain shape, finger position, size, center points palm, and margin extracting Important transforming them vector. accuracy proposed 97% using SIFT CNN, equal 94.8% nearly without SIFT. Finally, was tried tested group persons its effectiveness proven after considering their observations.

Language: Английский

Research on the application of augmented reality in English vocabulary teaching: A study on improving learning outcomes through interactive experience DOI

Shuang Zheng

Journal of Computational Methods in Sciences and Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: May 2, 2025

Augmented reality (AR) is revolutionizing the way we interact with information by blending physical and digital worlds to create immersive interactive environments. In context of English vocabulary learning, AR offers an innovative approach enhance engagement, comprehension, retention. This research aims improve teaching through AR-based applications, addressing challenges such as pedagogical depth, diverse learning styles, integration real-world contexts. The study focuses on design deployment application that incorporates gamified elements, visuals, real-time feedback facilitate students’ retention understanding vocabulary. Data collection methods include tracking user interactions, gathering responses, assessing performance activities. collected data underwent preprocessing, which involved cleaning normalization. Principal component analysis (PCA) was employed extract irrelevant features from processed data. improved weighted hybrid deep feedforward neural network (IWH-DFNN) utilized predict student outcomes these experiences. weights (IWH) applied optimize hyperparameters (DFNN), thereby increasing model’s predictive accuracy regarding performance. proposed IWH-DFNN model demonstrated superior in improving enhancing experience, achieving high recall (92.70%), precision (95%), (97%), F1-score (89%), minimal loss (0.03). findings suggest environments have potential integrating machine algorithms for adaptive within settings. creates a more engaging, customized, efficient environment.

Language: Английский

Citations

Advancing Human-Computer Interaction: AI-Driven Translation of American Sign Language to Nepali Using Convolutional Neural Networks and Text-to-Speech Conversion Application DOI