No abstract

As the global population grows, and urbanization becomes more prevalent, cities often struggle to provide convenient, secure, and sustainable lifestyles due to the lack of necessary smart technologies. Fortunately, the Internet of Things (IoT) has emerged as a solution to this challenge by connecting physical objects using electronics, sensors, software, and communication networks. This has transformed smart city infrastructures, introducing various technologies that enhance sustainability, productivity, and comfort for urban dwellers. By leveraging Artificial Intelligence (AI) to analyze the vast amount of IoT data available, new opportunities are emerging to design and manage futuristic smart cities. In this review article, we provide an overview of smart cities, defining their characteristics and exploring the architecture of IoT. A detailed analysis of various wireless communication technologies employed in smart city applications is presented, with extensive research conducted to determine the most appropriate communication technologies for specific use cases. The article also sheds light on different AI algorithms and their suitability for smart city applications. Furthermore, the integration of IoT and AI in smart city scenarios is discussed, emphasizing the potential contributions of 5G networks coupled with AI in advancing modern urban environments. This article contributes to the existing literature by highlighting the tremendous opportunities presented by integrating IoT and AI, paving the way for the development of smart cities that significantly enhance the quality of life for urban dwellers while promoting sustainability and productivity. By exploring the potential of IoT, AI, and their integration, this review article provides valuable insights into the future of smart cities, demonstrating how these technologies can positively impact urban environments and the well-being of their inhabitants.

Automatic identification of key points within objects is crucial in various application domains. This paper presents a novel framework for accurately estimating the key point within an object by leveraging deep neural network-based object detection. The proposed framework is built upon a training dataset annotated with four non-overlapping bounding boxes, one of which shares a coordinate with the key point. These bounding boxes collectively cover the entire object, enabling automatic annotation if region annotations around the key point exist. The trained object detector is then utilized to generate detection results, which are subsequently post-processed to estimate the key point. To validate the effectiveness of the framework, experiments were conducted using two distinct datasets: cross-sectional images of a parawood log and pupil images. The experimental results demonstrate that our proposed framework surpasses previously proposed approaches in terms of precision, recall, F1-score, and other domain-specific metrics. The improvement in performance can be attributed to the unique annotation strategy and the fusion of object detection and key point estimation within a unified deep learning framework. The contribution of this study lies in introducing a novel framework for closely estimating key points within objects based on deep neural network-based object detection. By leveraging annotated training data and post-processing techniques, our approach achieves superior performance compared to existing methods. This work fills a critical gap in the field by integrating object detection and key point estimation, which has received limited attention in previous research. Our framework provides valuable insights and advancements in key point estimation techniques, offering potential applications in precise object analysis and understanding. Doi: 10.28991/HIJ-2023-04-01-08 Full Text: PDF

Human activity recognition (HAR) attempts to classify performed activities from data retrieved from different sensors attached to the body. Most publications pertaining to HAR based on deep neural networks (DNNs) report the development of a suitable architecture to improve recognition accuracy by increasing the parameters of the architecture. Our work follows a different approach by attempting to optimise DNN-based HAR by reducing the dimensions of acceleration data, by finding a suitable sample size for processing by the DNN and by reducing the parameters of the proposed architecture. The experiments rely on employing two previously presented DNN-based HAR architectures as the baselines and starting points to create our candidate architectures. The variations in the dimensions of acceleration data, i.e., { xy, yz, xz, x, y, z }, and the sample size, i.e. {4, 6, 8} s duration, to these candidate architectures are experimented to produce the winner architecture which takes the shortest sample size and the minimal dimensions of acceleration data while preserving the recognition precision. The results indicate that despite the number of parameters is approximately half of the baseline architecture with two dimensions of acceleration data and shorter sample size (i.e., using a sample of 4 s duration instead of 8 s and only the xy axes of acceleration data), the resulting DNN-based HAR classifiers can produce comparable or better recognition precision than the baseline classifiers. The experimental results were obtained using three different popular datasets: the WISDM, the UCI HAR, and the Real World 2016. The proposed classifiers with optimised settings are useful as they require less processing time and reduce power consumption, both in terms of retrieving acceleration data from the sensor and the CPU processing time. Furthermore, they reduce the memory requirements for parameter storing and are suitable for incorporation in a wearable device.

To avoid problems related to a school bus service such as kidnapping, children being left in a bus for hours leading to fatality, etc., it is important to have a reliable transportation service to ensure students’ safety along journeys. This research presents a high accuracy child monitoring system for locating students if they are inside or outside a school bus using the Internet of Things (IoT) via Bluetooth Low Energy (BLE) which is suitable for a signal strength indication (RSSI) algorithm. The in/out-bus child tracking system alerts a driver to determine if there is a child left on the bus or not. Distance between devices is analyzed for decision making to affiliate the zone of the current children’s position. A simplified and high accuracy machine learning of least mean square (LMS) algorithm is used in this research with model-based RSSI localization techniques. The distance is calculated with the grid size of 0.5 m × 0.5 m similar in size to an actual seat of a school bus using two zones (inside or outside a school bus). The averaged signal strength is proposed for this research, rather than using the raw value of the signal strength in typical works, providing a robust position-tracking system with high accuracy while maintaining the simplicity of the classical trilateration method leading to precise classification of each student from each zone. The test was performed to validate the effectiveness of the proposed tracking strategy which precisely shows the positions of each student. The proposed method, therefore, can be applied for future autopilot school buses where students’ home locations can be securely stored in the system used for references to transport each student to their homes without a driver.