The increasing demands of several emergent services brought new communication problems to vehicular networks (VNs). It is predicted that the transmission system assimilated with unmanned aerial vehicles (UAVs) fulfills the requirement of next-generation vehicular network. Because of its higher flexible mobility, the UAV-aided vehicular network brings transformative and far-reaching benefits with extremely high data rates; considerably improved security and reliability; massive and hyper-fast wireless access; much greener, smarter, and longer 3D communications coverage. The clustering technique in UAV-aided VN is a difficult process because of the limited energy of UAVs, higher mobility, unstable links, and dynamic topology. Therefore, this study introduced an Enhanced Artificial Gorilla Troops Optimizer–based Clustering Protocol for a UAV-Assisted Intelligent Vehicular Network (EAGTOC-UIVN). The goal of the EAGTOC-UIVN technique lies in the clustering of the nodes in UAV-based VN to achieve maximum lifetime and energy efficiency. In the presented EAGTOC-UIVN technique, the EAGTO algorithm was primarily designed by the use of the circle chaotic mapping technique. Moreover, the EAGTOC-UIVN technique computes a fitness function with the inclusion of multiple parameters. To depict the improved performance of the EAGTOC-UIVN technique, a widespread simulation analysis was performed. The comparison study demonstrated the enhancements of the EAGTOC-UIVN technique over other recent approaches.

Human action and activity recognition are clues that alleviate human behavior analysis. Human action recognition (HAR) becomes a significant challenge in various applications involving human computer interaction (HCI) and intellectual video surveillance for enhancing security in distinct fields. Precise action recognition is highly challenging because of the variations in clutter, backgrounds, and viewpoint. The evaluation method depends on the proper extraction and learning of data. The achievement of deep learning (DL) models results in effectual performance in several image-related tasks. In this view, this paper presents a new quantum water strider algorithm with hybrid-deep-learning-based activity recognition (QWSA-HDLAR) model for HCI. The proposed QWSA-HDLAR technique mainly aims to recognize the different types of activities. To recognize activities, the QWSA-HDLAR model employs a deep-transfer-learning-based, neural-architectural-search-network (NASNet)-based feature extractor to generate feature vectors. In addition, the presented QWSA-HDLAR model exploits a QWSA-based hyperparameter tuning process to choose the hyperparameter values of the NASNet model optimally. Finally, the classification of human activities is carried out by the use of a hybrid convolutional neural network with a bidirectional recurrent neural network (HCNN-BiRNN) model. The experimental validation of the QWSA-HDLAR model is tested using two datasets, namely KTH and UCF Sports datasets. The experimental values reported the supremacy of the QWSA-HDLAR model over recent DL approaches.

Recently, consumer product recognition comprises harnessing cutting-edge technologies namely artificial intelligence (AI) and computer vision (CV) to develop the purchasing expedition. This technology allows retailers to utilize robust product recognition systems that precisely identify and categorize products in real time. The comprehension of automatic product identification becomes of major importance for both social and economic improvement since it is more reliable and time-consuming than manual function. Product detection through images is a complex task in the domain of CV. This can be obtained the improving consideration because of the excellent application viewpoint like visually impaired assistance, stock tracking, automatic checkout, and planogram compliance. Currently, deep learning (DL) prefers a successful progression with great achievements in object detection and image classification. Therefore, this study presents Advanced Consumer Product Recognition using the Aquila Optimization Algorithm with Deep Learning (ACPR-AOADL) technique. The proposed ACPR-AOADL model utilizes hyperparameter-tuned DL concepts for the identification of consumer products. To achieve this, the ACPR-AOADL model first pre-processes the input data utilizing a Wiener filter (WF) to improve the image quality. Besides, the YOLO-v8 model with a deep residual network (DRN) as a backbone network can be applied for the product detection process. For product classification, the deep belief network (DBN) approach can be used. To boost the complete product detection process, the ACPR-AOADL technique involves AOA based hyperparameter selection process. The performance analysis of the ACPR-AOADL method can be examined under the Product-10K dataset. Wide-ranging results stated that the ACPR-AOADL technique reaches enhanced classification performance over other compared approaches.

The use of computer-aided diagnostic (CAD) models has been proposed to aid in the detection and classification of breast cancer. In this work, we evaluated the performance of multilayer perceptron neural network and nonlinear support vector machine models to classify breast cancer nodules. From the contour of 569 samples, ten morphological features were used as input to the classifiers. The average results obtained in the set of 50 simulations performed show that the proposed models showed good performance (all exceeded 90.0%) in terms of accuracy in the test set. The nonlinear support vector machine algorithm stands out when compared to the proposed multilayer perceptron neural network algorithm, with 99% accuracy and a 2% false-negative rate. The neural network model presented lower performance than the nonlinear support vector machine classifier. With the application of the proposed models, the average results obtained are promising in the classification of breast cancer.

In recent years, the next-generation photonics and wearable devices have been progressing quickly towards the era of the Internet of Things (IoT) and artificial intelligence (AI), to reach a higher level of intelligence, comfort, convenience, and connection. AI technology is becoming increasingly important in the domain of tele-exercise, permitting workout programs to be customized to users’ requirements and capabilities. AI is applied to analyze data gathered from wearable electronics and environmental sensors and offer personalized recommendations and responses to the customers. Furthermore, a rise in the number of healthcare wearable IoT choices is making it challenging for individuals, healthcare specialists, and doctors to find the correct smart device that best suits their needs or treatments. This paper presents a Consumer Devices-Driven Smart Health Monitoring Framework Using Independent Component Analysis and Ensemble Classifiers (CDSHMF-ICAEC) model. The aim is to explore the integration of AI-driven consumer electronics and wearable IoT devices for enhanced user experience and real-time health monitoring. Initially, the inter-quartile range (IQR) method is applied in the data pre-processing stage for scaling the input data. Moreover, the independent component analysis (ICA) technique is utilized for the feature selection process to recognize and retain the most significant attributes from a dataset. Furthermore, the ensemble classifiers of variational autoencoder (VAE), temporal convolutional network (TCN), and deep belief network (DBN) models are employed for classification. The experimentation of the CDSHMF-ICAEC method is performed under the Wearables dataset. The comparison study of the CDSHMF-ICAEC method portrayed a superior accuracy value of 98.57% over existing techniques.

(LiDAR)-based 3-D mapping suffers from cumulative drift causing global misalignment, particularly in global navigation satellite system (GNSS)-constrained environments. To address this, we propose a unified framework that fuses LiDAR, GNSS, and inertial measurement unit (IMU) data for high-resolution city-scale mapping. The method performs velocity-based temporal alignment using dynamic time warping and refines GNSS and IMU signals via extended Kalman filtering. Local maps are built using normal distributions transform-based registration and pose graph optimization with loop closure detection, while global consistency is enforced using GNSS-constrained anchors followed by fine registration of overlapping segments. We also introduce a large-scale multimodal dataset captured in Perth, Western Australia to facilitate future research in this direction. Our dataset comprises 144 000 frames acquired with a 128-channel Ouster LiDAR, synchronized RTK–GNSS trajectories, and MEMS–IMU measurements across 21 urban loops. To assess geometric consistency, we evaluated our method using alignment metrics based on road centerlines and intersections to capture both global and local accuracy. The proposed framework reduces the average global alignment error from 3.32 to 1.24 m, achieving a 61.4% improvement, and significantly decreases the intersection centroid offset from 13.22 to 2.01 m, corresponding to an 84.8% enhancement. The constructed high-fidelity map and raw dataset are publicly available through IEEE Dataport and its visualization can be viewed in the provided Demo. This dataset and method together establish a new benchmark for evaluating 3D city mapping in GNSS-constrained environments, with source code available at GitHub Repository.