Evidence from psychology and behaviour therapy shows that engaging in sports activities at home might help alleviate stress and depression during COVID-19 lockdown periods. A clever virtual coach that provides table tennis instruction at a low cost without invading privacy might be a great way to maintain a healthy lifestyle without leaving the house. In this article, we look at creating the second main constituent of the virtual-coach table tennis shadow-play training scheme: an evaluation system for the effectiveness of the forehand stroke. This research was carried out to demonstrate the efficacy of the suggested bidirectional long-short-term memory (BLSTM) model in assessing the table tennis forehand shadow-play sensory data supplied by the authors in comparison with LSTM timeseries investigation approaches. Information was collected by tracking the rackets of 16 players as they performed forehand strokes and assigning assessment ratings to each stroke based on the input of three instructors. The scientists looked at how the hyperparameter values, which are chosen via an optimisation approach, affected the behaviour of DL models. The adaptive learning differential approach has been introduced to enhance the functionality of the standard dragonfly algorithm. Optimal BLSTM settings are selected with the help of the enhanced dragonfly algorithm (IDFOA). Evaluation of Tennis Teaching Effect Using Optimized DL Model with Cloud Computing System Volume 16 (2024), Issue 2 17 The experimental findings of this study indicate that the BLSTM-IDFOA is the most effective regression approach currently available.

In the quest for autonomous vehicle safety and road infrastructure management, traffic sign recognition (TSR) remains paramount. Recent advancements in accuracy across various benchmarks have been identified in the literature concerning this essential task. Such technology might remain absent in older vehicles, while integration into Advanced Driver Assistance Systems (ADAS) is common in more recent models. Yet, the capability of these systems to function proficiently under diverse driving conditions has not been widely investigated. A framework has been devised to allow a moving vehicle to detect traffic signs, targeting the enhancement of driver safety and the diminishment of accidents. The present research introduces an innovative methodology, amalgamating the extreme learning machine (ELM) method with deep-learning paradigms, in response to experimental discoveries. As a pioneering computational approach in neural network-based learning, ELM facilitates rapid training and commendable generalization. An accuracy of 95.00% was achieved by the proposed model. By utilizing the Horse Herd Optimization method (HHOA), the memory consumption is minimized in the more sophisticated approach of stacked ELM (SELM) within the deep-learning framework. This study contributes to the understanding of potential challenges that may be encountered during TSR tasks, and lays the groundwork for future investigation by proffering a diverse set of evaluations for various road scenarios. Consistency in the utilization of professional terms is maintained throughout.

Glaucoma is an increasing ocular disease and one of the major causes of irreversible blindness globally. Thus, it is of utmost importance to diagnose glaucoma accurately and in a timely manner for appropriate clinical intervention. Retinal fundus image analysis is a non-invasive approach for glaucoma screening. However, it is a tedious and subjective process for accurate glaucoma detection. In this work, a customized convolutional neural network (CNN) model is proposed for glaucoma classification using fundus images, and an attention mechanism is employed for improved discriminative feature learning. Experiments were carried out on two publicly available datasets, namely Drishti-GS1 and ACRIMA, for unbalanced and balanced datasets, respectively. Data augmentation and hyperparameter tuning were employed for improved model generalization. To increase model explainability and trust for accurate glaucoma screening, gradient-weighted class activation mapping (Grad-CAM) is employed for accurate glaucoma screening. For accurate glaucoma screening, the proposed model attained 90.32% and 96.45% accuracy on the Drishti-GS1 and ACRIMA datasets, respectively, with higher sensitivity and AUC scores. Thus, it is evident that the proposed model employing attention mechanisms and explainable AI attained higher accuracy for accurate glaucoma screening.