Three-Dimensional Simulation for Training Autonomous Vehicles in Smart City Environments

Real-time object detection and tracking in video streams has become a fundamental component of modern intelligent systems, enabling applications in surveillance, autonomous driving, robotics, healthcare monitoring, and smart cities. The rapid advancement of deep learning techniques has significantly improved the accuracy and efficiency of detecting and tracking multiple objects under complex and dynamic environments. This study presents a deep learning–based framework for real-time object detection and tracking in continuous video streams, emphasizing both computational efficiency and detection robustness.The proposed system integrates a state-of-the-art convolutional neural network (CNN)-based object detector with a reliable multi-object tracking algorithm. The detection module identifies objects of interest in each frame, while the tracking component maintains consistent identities across consecutive frames by utilizing motion prediction and feature association strategies. To achieve real-time performance, the model is optimized through lightweight network architectures and parallel processing on GPU hardware. Data augmentation and transfer learning techniques are employed to enhance generalization and reduce training time. Performance evaluation is conducted on benchmark video datasets using standard metrics such as precision, recall, mean Average Precision (mAP), and tracking accuracy. Experimental results demonstrate that the proposed framework achieves high detection accuracy while maintaining low latency, making it suitable for real-time deployment. Furthermore, the system effectively handles challenges such as occlusion, scale variation, and illumination changes. This research presents a comprehensive study on real-time object detection and tracking in video streams using deep learning techniques. The proposed framework integrates YOLO-based object detection with Deep SORT multi-object tracking to achieve robust spatial localization and temporal identity preservation. The study evaluates performance using benchmark datasets including COCO, MOT17, and KITTI. Mathematical formulations of detection loss, Kalman filtering, and data association are presented. Experimental analysis using precision, recall, mAP, MOTA, ROC-AUC, and PR-AUC demonstrates that the system achieves high detection accuracy while maintaining real-time performance above 35 frames per second. The findings confirm that deep learning-based frameworks provide scalable and efficient solutions for surveillance, autonomous vehicles, and intelligent monitoring systems.

Many cities around the world have embarked on ambitious programmes towards creating Smart Cities where information, diverse digital opportunities, and collective intelligence can be harnessed ubiquitously. Smart Cities are conceptualized using citywide smart and intelligent architectures informed by the context in which they are implemented. These architectures make it possible to access information and intelligence anywhere and at any time. Information processing and computing is embedded within the urban infrastructures to a point where immovable city entities such as traffic lights are more intelligent to make real-time decisions based on the happenings in the environment in which they are deployed. Advanced development of ambient computing within the realm of Smart Cities will further culminate into possibilities such as vehicle-to-vehicle communication (V2V) and mobile-to-mobile (M2M) communication. Using extensive and critical literature review, this chapter specifically focusses on the design of information architectures that will ultimately support the enshrining of spatial intelligence within Smart City environments hinged on the internet of things (IoT) and cloud/fog computing. The chapter presents latest trends in the research and practice of ambient intelligence (AmI) linked to the realization of the key principles of Smart Cities from the information topology and architecture point of view. A conceptual ambient intelligence architecture that highlights the building blocks of any ambient intelligence architecture as deployed in Smart City environments is also proposed. The proposed conceptual architecture can be used as a blueprint in the design of ambient intelligence topologies and architectures in different contextual settings.

With the popularization of Internet-of-things (IoT) and wireless communication systems, a diverse set of applications in smart cities are emerging to improve the city-life. These applications usually require a large coverage area and minimal operation and maintenance cost. To this end, the recently emerging backscatter communication (BC) is gaining interest in both industry and academia as a new communication paradigm that provides high energy efficient communications that may even work in a battery-less mode and, thus, it is well suited for smart city applications. However, the coverage of BC in urban area deployments is not available, and the feasibility of its utilization for smart city applications is not known. In this article, we present a comprehensive coverage study of a practical cellular carrier-based BC system for indoor and outdoor scenarios in a downtown area of a Helsinki city. In particular, we evaluate the coverage outage performance of different low-power and wide area technologies, i.e., long range (LoRa) backscatter, arrow band-Internet of Things (NB-IoT), and Bluetooth low energy (BLE) based BC at different frequencies of operation. To do so, we carry out a comprehensive campaign of simulations while using a sophisticated three-dimensional (3D) ray tracing (RT) tool, ITU outdoor model, and 3rd generation partnership project (3GPP) indoor hotspot model. This study also covers the energy harvesting aspects of backscatter device, and it highlights the importance of future backscatter devices with high energy harvesting efficiency. The simulation results and discussion provided in this article will be helpful in understanding the coverage aspects of practical backscatter communication system in a smart city environment.

The proliferation of Internet of Things (IoT) devices has opened new roads for collaborative distributed applications, particularly in smart city environments, where a variety of resources, including sensing, actuation, computation, and storage, are essential for providing effective location-based services. This paper specifically focuses on the sharing of heterogeneous resources among IoT applications in smart cities. By leveraging game-theoretic principles, this study addresses resource allocation through a combinatorial double auction. The solution is rooted in the concept of Social IoT (SIoT), where Internet-connected objects create dynamic social networks based on rules set by their owners. Social relationships, such as ownership and co-location, are leveraged to form groups offering enhanced reliability resource bundles. The proposed solution offers several key economic properties, including incentive compatibility, individual rationality, and a balanced budget, while maintaining low computational complexity. Simulation results demonstrate that the proposed combinatorial double auction mechanism achieves over 70% successful resource allocation for up to 1000 requests, maintains computational efficiency with execution times under 30 s, and ensures economic properties such as incentive compatibility and individual rationality, making it a scalable and practical solution for large-scale smart city IoT applications.

Smart cities have expanded rapidly with the introduction of Internet of Things (IoT) devices and edge computing techniques. Such technologies have transformed how cities function through the ability to deliver smart services and analyze data in real time. But that advancement also adds to the demanding issues about privacy and data security. In smart city environments, privacy-preserving is essential to protect sensitive data collected from different channels against leakage, abuse and unauthorized access. This work is a review of data security and privacy-preserving models made for edge computing and smart cities. It examines the advantages and disadvantages of existing machine learning, blockchain, cryptography, hybrid architectures of optimizing resources, scalability and efficiency. The review discusses open challenges, including interoperability, resource constraints and governance issues, as well as future research directions such as quantum-secure security, federated learning and AI-based adaptive models. The concepts aim to help researchers and policymakers establish strong, durable smart city infrastructures.

Abstract: Radio-over-Free-Space Optics (RoFSO) systems have emerged as a promising solution to address the growing demand for high-capacity and flexible wireless communications in smart city environments. RIS-assisted RoFSO systems offer a compelling solution to the communication challenges inherent in smart city applications. By leveraging the reflective and reconfigurable capabilities of intelligent surfaces, these systems can overcome the traditional LOS limitations and significantly mitigate the adverse effects of atmospheric turbulence, as accurately modeled by the Málaga-M distribution. The demonstrated improvements in outage probability, channel capacity, and bit error rate highlight their potential to provide reliable, high-speed, and flexible wireless connectivity in dense urban environments. However, the performance of RoFSO links is significantly affected by atmospheric turbulence. The Málaga-M distribution has gained prominence as a comprehensive model to capture a wide range of turbulence conditions. Recent advancements in Reconfigurable Intelligent Surfaces (RISs) offer a new dimension to mitigating the limitations of FSO channels. This paper provides a comprehensive literature review on the integration of RIS with RoFSO systems over Málaga-M fading, focusing on their applicability in smart city deployments.

Smart cities have the ability to transform data from the physical world into data in the cyber world and monitor it in real time, and contribute to intelligent services for users in terms of entertainment, transportation, health care, the environment and energy. However, concerns arise about security, because smart city apps not only collect sensitive information from people and their environments, but also affect people's lives. One of the main challenges on the smart city application is privacy where the smart city infrastructure enables to monitoring and surveillance a massive data that can sharing, processing, generating and analyzing large amounts of personal data such as their geo location, characteristics, activities and if it links this data together can created profiles or people that against their privacy right.

Use of autonomous vehicles aims to eventually reduce the number of motor vehicle fatalities caused by humans. Deep learning plays an important role in making this possible because it can leverage the huge amount of training data that comes from autonomous car sensors. Automatic recognition of traffic light and vehicle signal is a perception module critical to autonomous vehicles because a deadly car accident could happen if a vehicle fails to follow traffic lights or vehicle signals. A practical Traffic Light Recognition (TLR) or Vehicle Signal Recognition (VSR) faces some challenges, including varying illumination conditions, false positives and long computation time. In this chapter, we propose a novel approach to recognize Traffic Light (TL) and Vehicle Signal (VS) with high dynamic range imaging and deep learning in real-time. Different from existing approaches which use only bright images, we use both high exposure/bright and low exposure/dark images provided by a high dynamic range camera. TL candidates can be detected robustly from low exposure/dark frames because they have a clean dark background. The TL candidates on the consecutive high exposure/bright frames are then classified accurately using a convolutional neural network. The dual-channel mechanism can achieve promising results because it uses undistorted color and shape information of low exposure/dark frames as well as rich texture of high exposure/bright frames. Furthermore, the TLR performance is boosted by incorporating a temporal trajectory tracking method. To speed up the process, a region of interest is generated to reduce the search regions for the TL candidates. The experimental results on a large dual-channel database have shown that our dual-channel approach outperforms the state of the art which uses only bright images. Encouraged by the promising performance of the TLR, we extend the dual-channel approach to vehicle signal recognition. The algorithm reported in this chapter has been integrated into our autonomous vehicle via Data Distribute Service (DDS) and works robustly in real roads.

A lot of research work has been carried out on the smart city and its applications in last two decades. Smart transportation is an application of smart city and also an important factor in building smart cities. Due to the complexity in passenger's traffic in big cities, a smart transportation system is needed to resolve the issue. This paper examines the concept of a vision-based transportation system to resolve the complexities in passenger's traffic and improving the productivity of existing transportation system. The vision-based transportation systems are expected to have positive mobility effects and be able to subsidize to the green environment by improving air quality and energy conservation for planning and developing a smart city. Thus, vision-based transportation is possibly an important approach for supporting and improving the transportation system. Vision-based transportation is a set of technologies applied to transportation infrastructure and vehicles to improve their performance. More specifically, vision-based transportation involves the application of established communications, controls, electronics, computer hardware, and software technologies in the transportation system. The paper examines in detail a broad range of vision-based transportation technologies and the expected benefits. These benefits include improved transportation system, travel time, throughput, cost savings, and most importantly safety and security. The paper considers a set of barriers that are required for the improvement of autonomous vehicles.

Accurate segmentation and delineation of the sub-tumor regions are very challenging tasks due to the nature of the tumor. Traditionally, convolutional neural networks (CNNs) have succeeded in achieving most promising performance for the segmentation of brain tumor; however, handcrafted features remain very important in identification of tumor's boundary regions accurately. The present work proposes a robust deep learning-based model with three different CNN architectures along with pre-defined handcrafted features for brain tumor segmentation, mainly to find out more prominent boundaries of the core and enhanced tumor regions. Generally, automatic CNN architecture does not use the pre-defined handcrafted features because it extracts the features automatically. In this present work, several pre-defined handcrafted features are computed from four MRI modalities (T2, FLAIR, T1c, and T1) with the help of additional handcrafted masks according to user interest and fed to the convolutional features (automatic features) to improve the overall performance of the proposed CNN model for tumor segmentation. Multi-pathway CNN is explored in this present work along with single-pathway CNN, which extracts simultaneously both local and global features to identify the accurate sub-regions of the tumor with the help of handcrafted features. The present work uses a cascaded CNN architecture, where the outcome of a CNN is considered as an additional input information to next subsequent CNNs. To extract the handcrafted features, convolutional operation was applied on the four MRI modalities with the help of several pre-defined masks to produce a predefined set of handcrafted features. The present work also investigates the usefulness of intensity normalization and data augmentation in pre-processing stage in order to handle the difficulties related to the imbalance of tumor labels. The proposed method is experimented on the BraST 2018 datasets and achieved promising results than the existing (currently published) methods with respect to different metrics such as specificity, sensitivity, and dice similarity coefficient (DSC) for complete, core, and enhanced tumor regions. Quantitatively, a notable gain is achieved around the boundaries of the sub-tumor regions using the proposed two-pathway CNN along with the handcrafted features. Graphical Abstract This data is mandatory. Please provide.

In this developing paradigm, heterogeneous network infrastructure is combined with ubiquitous devices for sensing, smart control systems and big data analytics to create “smart cities.” Enhancing the overall quality of life is one of their main goals and they do so through smart services they offer in a broad range of areas such as transportation, healthcare, entertainment (including virtual reality), the environment (including pollution) and energy. To be able to offer those services, the job of big data and their analysis is complicated, since it allows for the extraction of useful insights from the enormous amounts of data that the smart cities produce. The current state-of-the-art research efforts aimed at big data analytics in a smart city setting are investigated and presented in this chapter. In particular, a comprehensive literature review on the big data analytic frameworks for smart cities is provided. Following that, a short talk on IoT and big data is briefed. Owing to different data that the smart cities collect, which frequently have contradictory processing needs, appropriate investigative methods based on the data type are also recommended, depending on the data type. The many smart city applications were investigated and a perspective on the difficulties associated with smart city development was developed. In this chapter, the methods addressing the smart city issues and development are examined, and the possibility for the employment of integrated IoT and big data analytics is presented. Finally, as a case study, a briefing on a call service application for the fire department in a smart city is provided to round up the presentation.

The government announced the fourth comprehensive plan for smart cities in May, following the enactment of the world's first smart city-related law in 2008 and the implementation of smart city policies.Although it has led to the development of smart infrastructure that integrates information and communication technology into new cities, the government's top-down approach has also faced criticism.Recently, the significance of the Smart City Living Lab, involving citizen participation, has gained prominence in the evolving landscape of smart city policies.Initiatives like the smart city challenge and integrated platform supply projects underscore the shift toward citizen engagement.The government is now advocating for a smart city convergence course tailored to university characteristics to nurture innovative talents for smart cities.Domestic universities have started to embrace living labs in response to this trend.This study proposes a policy direction for the efficient operation of the Smart City Living Lab project, which local governments increasingly adopt.It also delves into the management of the Resident-Participatory Living Lab, which is university-led and involves citizen participation and its implications.This study of living labs in Dangjin, this research explores strategies to activate and manage living labs at domestic universities through a case study of living labs in Dangjin city.

Protect your data and I’ll rank its utility: A framework for utility analysis of anonymized mobility data for smart city applications

Internet-of-Things (IoT) technologies in the past decade have matured both in the hardware and software aspects for large-scale deployment. Alongst IoT, the Smart Cities Concept is also taking shape. Pilot projects and implementations in multiple cities are trying to find out the feasibility and applicability of Smart City Information and Communications Technology (ICT). IoT assets along with the legacy assets are essential for Smart City ICT implementations. With the evolution of Smart Cities and concentration of people in the cities, it becomes necessary to be ready for future Humanitarian Assistance and Disaster Recovery (HADR) operations. But the huge void in heterogeneous IoT and legacy technologies create a big hurdle in establishing and handling the HADR operations. This aim of this PhD is to investigate the interoperability aspects amongst the various IoT technologies and Smart City concepts. The goal is to create a framework and an architecture for allowing the interoperable operation of ICT assets in a Smart City environment. This framework would enable rapid deployment of HADR relevant technology assets on the ground allowing multiple HADR agencies to seamlessly communicate while having shared Situational Awareness (SA) and complementing each others capabilities.

Abstract. The smart cities that promise a sustainable future cannot be thought of independently from the spatial information infrastructure. It is very important to keep the spatial data infrastructure up-to-date for guidance and information in smart city applications (SCA). Easy and low-cost acquisition is an important factor in updating spatial data. Today, there are many measurement techniques to collecting 3-D spatial data of urban areas and land topography. On the other hand, indoor measurement and 3-D modelling techniques are used in the creation of building information modelling (BIM). In this study, measurement techniques that provide 3-D point cloud data to SCA are examined. Consequently, image based photogrammetry and dense matching methods enable low cost measurement than LiDAR based active measurement. The active 3-D measurement techniques have high accuracy especially for mid and long ranges. The LiDAR, that can be applied at day or night time, offer more opportunity to performing SCA like autonomous vehicle and robotic navigations. Nevertheless, LiDAR can only capture structure, not texture, and therefore has limits to the types of data that it can capture. The LiDAR and image based methods are complement to each other in 3-D reality capture. The 3-D measurement techniques are exploited according to SCA as alone or together.