Smart Objects Research Articles

Smart-YOLO glass: Real-time video based obstacle detection using paddling/paddling SAB YOLO network1

Visual impairment people have many difficulties in everyday life, including communicating and getting information, as well as navigating independently and safely. Using auditory alerts, our study hopes to improve the lives of visually impaired individuals by alerting them to items in their path. In this research, a Video-based Smart object detection model named Smart YOLO Glass has been proposed for visually impaired persons. A Paddling - Paddling Squeeze and Attention YOLO Network model is trained with multiple images to detect outdoor objects to assist visually impaired people. In order to calculate the distance between a blind person and obstacles when moving from one location to another, the proposed method additionally included a distance-measuring sensor. The visually impaired will benefit from this system’s information about around objects and assistance with independent navigation. Recall, accuracy, specificity, precision, and F-measure were among the metrics used to evaluate the proposed strategy. Because there is less time complexity, the user can see the surrounding environment in real time. When comparing the proposed technique to Med glasses, DL smart glass, and DL-FDS, the total accuracy is improved by 7.6%, 4.8%, and 3.1%, respectively.

Trinal Optics for Blind People Using Smart Object Recognition for Navigation Through Voice Assistant

The project Trinal Optics is a wearable device, it will be able to make blind people to live their lives more independently. The device is combination of hardware, software and Technologies such as Camera, Arduino Uno, Voice Assistant, Bluetooth Headset/Earphone, Ultrasonic Sensor, Object Detection, and Distance Analysis. These components help user to understand the environment better. The Camera is used to capture the image for object detection and then Voice assistant is conveying audio message to the user along with the Distance analysis of object, Arduino Uno is used to integrate hardware and software, navigation is used to guide the users to its destination Ultrasonic sensor is used to make the beep sound if there is any object in front of the user. Trinal Optics looks forward to help blind individuals to have a bit easy life with the help of technology. Keywords: Arduino Uno, Camera, Bluetooth Earphone, Voice Assistant, Object Detection, Distances Analysis, Ultrasonic Sensor.

Interrelations between Printing Patterns and Residual Stress in Fused Deposition Modelling for the 4D Printing of Acrylonitrile Butadiene Styrene and Wood–Plastic Composites

Four dimensional printing enables the advanced manufacturing of smart objects that can morph and adapt shape over time in response to stimuli such as heat. This study presents a single-material 4D printing workflow which explores the residual stress and anisotropy arising from the fused deposition modelling (FDM) printing process to create heat-triggered self-morphing objects. In particular, the study first investigates the effect of printing patterns on the residual stress of FDM-printed acrylonitrile butadiene styrene (ABS) products. Through finite element analysis, the raster angle of printing patterns was identified as the key parameter influencing the distribution of residual stresses. Experimental investigations further reveal that the non-uniform distribution of residual stress results in the anisotropic thermal deformation of printed materials. Thus, through the design of printing patterns, FDM-printed materials can be programmed with desired built-in residual stresses and anisotropic behaviours for initiating and controlling the transformation of 4D-printed objects. Using the proposed approach, any desktop FDM printers can be turned into 4D printers to create smart objects that can self-morph into target geometries. A series of 4D printing prototypes manufactured from conventional ABS 3D printing feedstock are tested to illustrate the use and reliability of this new workflow. Additionally, the custom-made wood–plastic composite (WPC) feedstocks are explored in this study to demonstrate the transposability of the 4D printing approach.

A miniaturized array microneedle tactile sensor for intelligent object recognition

Tactile sensors can provide intuitive information about the shape, size, and texture of objects, which helps robots make more accurate decisions. In this paper, based on the principle of triboelectrification and electrostatic induction, a miniature tactile sensor (MTS) for smart object recognition is designed using composite materials based on silica gel. Its diameter is only 6 mm, and the electrodes are arranged in a circumferential array, which has excellent shape adaptability and can recognize curved materials with curvature of 1/15 mm−1. By optimizing the sensor structure and process to obtain a microstructure with a spike-like shape, the sensitivity of the sensor is successfully improved, and the smallest and accurately detected surface material side length is 5 mm. In addition, the article provides a smart sensor system that can enhance the recognition function of robots with the help of machine learning technology, and the discrimination accuracy reaches 95.7%, which lays a solid foundation for practical applications.

Incorporating smart objects into BIM: A comprehensive review

Incorporating smart objects into BIM: A comprehensive review

Intelligent resource management in 5G/6G network by adopting edge intelligence for higher education systems

The Internet of Things (IoT) is expected to have a positive impact on various sectors, such as manufacturing, users, and operators. Due to the capabilities of smart objects to share their contents and interact with their social networks, the combined Social Internet of Educational Things (SIoET) for smart education is expected to provide users with more efficient services. The primary objective of this paper is to provide an efficient resource management by adopting edge intelligence technique for higher education system workloads on next-generation networks. As the number of tasks handled by these interconnected devices continues to rise, there is an escalating concern regarding resource competition and power consumption within the network. To mitigate these challenges, we propose an innovative edge intelligence-based resource allocation approach to intelligently allocate resources while minimizing latency and optimizing power consumption with increasing task volumes. The problem is dissected into two core sub-problems: computation resource optimization and power optimization. Leveraging the concept of Lagrangian and Bisection methods, our proposed algorithm, implemented within a Device-to-Device (D2D) -assisted edge intelligence network, demonstrates remarkable performance improvements. Network latency and energy consumption is reduced by up to 57.9 % and 82.3 % compared to the benchmark schemes. These findings hold promising implications for the integration of this work into higher education systems driven by SIoET. Furthermore, simulations indicate that the system maintains robust performance even in dynamic urban environments characterized by random device movement.

4D Fabrication of Two-Way Shape Memory Polymeric Composites by Electrospinning and Melt Electrowriting.

This work presents a new method for 4D fabrication of two-way shape memory materials that are capable of reversible shapeshifting right after manufacturing, upon application of proper heating and cooling cycles. The innovative solution presented here consists in the combination of highly stretched electrospun shape memory polymer (SMP) nanofibers with a melt electrowritten elastomer. More specifically, the stretched nanofibers are made of a biocompatible thermoplastic polyurethane (TPU) with crystallizable soft segments, undergoing melt-induced contraction and crystallization-induced elongation upon heating and cooling, respectively. Reversible actuation during crystallization becomes possible due to the elastic recovery of the elastomer component, obtained by melt electrowriting of a commercial TPU filament. Thanks to the design freedom offered by additive manufacturing, the elastomer structure also has the role of guiding the shape transformation. Electrospinning and melt electrowriting process parameters are set up so to obtain smart 4D objects capable of two-way shape memory effect (SME), and the possibility of reversible and repeatable actuation is demonstrated. The two components are then combined in different proportions with the aim of tailoring the two-way SME, taking into account the effect of design parameters such as the SMP content, the elastomer pattern, and the composite thickness.

O/S Level Interrupt Prediction for Performance and Energy Management on Android

Billions smartphones and smart objects battery-powered use Android, i.e. on the Linux kernel. To save energy, the main kernel leverage is to put processors in a low power state as soon as they are idle. It predicts the next event to estimate the sleep duration and choose a sleep state accordingly. Several wake-up sources (interrupts, events...) impact this prediction which is usually done considering them as a single source. The resulting signal is nearly random and difficult to predict. Processors are recently supporting deeper idle states but the prediction paradigm was never challenged. We propose to predict the next event by splitting the wake-up source signal into simpler event patterns. We describe a fast and efficient algorithm and its kernel-level performance evaluation. We compare our approach with multiple reference sleep state selection algorithms on actual ARM and x86 boards using classical mobile workloads. Our proposal detects correctly (up to 20% improved correctness leading to 5% reduced energy consumption) the time of next interrupt, and thus the right sleep level for the processor. We show and discuss the energy impact of the tested prediction algorithm and we compare it with the different generations of sleep level managers in the Linux kernel.

Cyber Attack Detection Using Machine Learning Techniques in IoT Networks

The Internet of Things is a rapidly emerging technology (IoT). Thanks to numerous sensors, billions of smart objects (referred to as "Things") may acquire data about themselves and their environment. So they may regulate and monitor industrial services or increase commercial services and activities. However, the Internet of Things is now more vulnerable than ever. Machine Learning (ML) has advanced significantly, bringing up new research avenues to solve current and future IoT challenges. On the other hand, machine learning is an effective method for identifying peril in intelligent devices and networks. Following a thorough literature review on Machine Learning methods and the necessity of IoT security, this study will assess numerous ML algorithms for threat detection and the various security methods, which are associated with Machine Learning techniques.

Sewage water management and healthcare monitoring in IoT using Optimized deep residual network

ABSTRACT The Internet of Things (IoT) is termed as the interconnection of different smart objects with respect to devices. In this research, two different application scenarios are considered to show the efficiency of the Deep Residual Network (DRN) through multicast routing. The entities involved in the process are IoT nodes, IoT heads, and base stations (BS). The nodes are allowed to capture the information, and the collected data are routed to BS through the head node. The process of routing is made using the CrowWhale optimisation algorithm that enables to transfer the data packets from IoT nodes to BS. In the sewage water management system, entering sewage water into fresh water is detected by DRN which is trained using an optimisation algorithm. In the healthcare system, heart disease prediction is done using DRN to detect normal and abnormal cases more effectively. The adopted CrowWhale-ETR+DRN offered energy, accuracy and sensitivity as 82.54, 0.967, and 0.978 with 100 nodes for the environmental protection dataset. The energy, accuracy, and sensitivity obtained by the proposed model are 83.232, 0.964, and 0.974 using 100 nodes for the heart disease dataset, respectively.

A New WRR Algorithm for an Efficient Load Balancing System in IoT Networks under SDN

The Internet of Things (IoT) connects various smart objects and manages a vast network using diverse technologies, which present numerous challenges. Software-defined networking (SDN) is a system that addresses the challenges of traditional networks and ensures the centralized configuration of network entities to manage network integrity. Furthermore, the uneven distribution of IoT network load results in the depletion of IoT device resources. To address this issue, traffic must be distributed equally, requiring efficient load balancing to be ensured. This requires the development of an efficient architecture for IoT networks. The main goal of this paper is to propose a novel architecture that leverages the potential of SDN, the clustering technique, and a new weighted round-robin (N-WRR) protocol. The objective of this architecture is to achieve load balancing, which is a crucial aspect in the development of IoT networks as it ensures the network’s efficiency. Furthermore, to prevent network congestion and ensure efficient data flow by redistributing traffic from overloaded paths to less burdened ones. The simulation results demonstrate that our N-WRR algorithm achieves highly efficient load balancing compared to the simple weighted round-robin (WRR), and without the application of any load balancing method. Furthermore, our proposed approach enhances throughput, data transfer, and bandwidth availability. This results in an increase in processed requests.

Smart Objects and Replicas: A Survey of Tangible and Embodied Interactions in Museums and Cultural Heritage Sites

Tangible and embodied interactions are areas of research within HCI and interaction design. They refer to a way in which interacting with computer systems is closer to the way we interact with the real world. Instead of using devices such as a computer or a phone, we interact using seemingly non-technological objects, by moving our bodies or by using gestures. Since the early 2000s, tangible and embodied interactions have been applied and researched also in cultural heritage and museums, in an attempt to overcome issues induced by screen-based devices that may disengage visitors from the objects, their materiality and the physicality of the visit. This article surveys tangible and embodied interactions in museums, over a period of two decades since 2000. More than 120 projects have been researched and analysed thematically to provide a categorization based on cultural communication, interaction features and museological aspects. This categorization offers a conceptualization of tangible and embodied interactions in museums and cultural heritage; it suggests a terminology to describe the design characteristics of tangible and embodied interaction interventions, therefore facilitating the orientation of future research efforts in the field.

Complexity Induced Risks Posed by Upcoming Digital Ultra-Platforms: An Agent-Based Simulation Model

The global deployment of new digital technologies, such as fifth-generation broadband, Internet of Things, smart objects, and AI, is paving the way for new high-performance digital platforms. These platforms are capable of connecting both people and “things” at high speeds and near-zero latency, taking on the nature of digital “ultra-platforms”. However, the increased reach (i.e., more agents interacting across the platform) and responsiveness (i.e., more opportunities for agents to respond and adapt to each, as they interact) of digital “ultra-platforms” create an unprecedented level of complexity. In this article, we use an agent-based simulation model to study the different types of risks that arise from this increased level of complexity. Three different risks are identified that can hinder a platform's ability to create value. We refer to these risks as the digital black swan, the stacked deck, and push back. Each risk is considered and analyzed from the perspective of specific stakeholders who may be affected by platform externalities. These include the platform owner, platform users, and others who may have no direct interest in the platform. In this way, the article aims to contribute to the literature on digital platform management by focusing on challenges that are emerging from new digital technologies yet have thus far received little attention from either scholars or managers.

AIWLO-WMO Model Based an Artificial Intelligence to Smart Waste Management for Smart Cities Environment

Smart waste management is one of the world's massive challenges, either in the advanced or started economies. Regularly, the technologies of artificial intelligence obtained recognition in presenting different computer techniques to solving smart waste challenge. The smart waste management of specified problems, events and doubts and partial statistics were capable for AI. Although this task did operate lots of findings, very few assessments proved the impact of AI to determine numerous difficulties of intelligent organization of waste. Perfect assessment of waste amount and condition is explained to Smart waste management technique development and model. Currently, heavy quantity of waste resources has become substantial enhanced with increasing population. Appropriate management of waste resources becomes necessary to decrease environmental degradation and recover value of life in smart cities. Smart Waste management supports to gather and heal waste resources from society. Suitable arrangement of waste objects needs the model of automated waste category models based on artificial intelligence and image processing-based methods. The aim of this paper, an automated artificial intelligence with world larva optimization supports waste management and organization (AIWLO-WMO) model is proposed for smart cities environment. The proposed model mainly develops a RetinaNet built object finding segment to recognize the presence of smart waste objects in the views. To better the category operation, Adagrad optimizer has used. To confirming the invented outcome of the AIWLO-WMO method, complete testing is performed on normal dataset and the found rates implied the authority of AIWLO-WMO model throughout other methods with expanded accuracy of 99.62%

Functional, hedonic, and social motivated consumer innovativeness as a driver of word-of-mouth in smart object early adoptions: an empirical examination in two product categories

Functional, hedonic, and social motivated consumer innovativeness as a driver of word-of-mouth in smart object early adoptions: an empirical examination in two product categories

YOLO glass: video-based smart object detection using squeeze and attention YOLO network

YOLO glass: video-based smart object detection using squeeze and attention YOLO network

SHPIA 2.0: An Easily Scalable, Low-Cost, Multi-purpose Smart Home Platform for Intelligent Applications

Sensors, electronic devices, and smart systems have invaded the market and our daily lives. As a result, their utility in smart home contexts to improve the quality of life, especially for the elderly and people with special needs, is getting stronger and stronger. Therefore, many systems based on smart applications and intelligent devices have been developed, for example, to monitor people’s environmental contexts, help in daily-life activities, and analyze their health status. However, most existing solutions have drawbacks related to accessibility and usability. They tend to be expensive and lack generality and interoperability. These solutions are not easily scalable and are typically designed for specific constrained scenarios. This paper tackles such drawbacks by presenting SHPIA 2.0, an easily scalable, low-cost, multi-purpose smart home platform for intelligent applications. It leverages low-cost Bluetooth Low Energy (BLE) devices featuring both BLE connected and BLE broadcast modes, to transform common objects of daily life into smart objects. Moreover, SHPIA 2.0 allows the collection and automatic labeling of different data types to provide indoor monitoring and assistance. Specifically, SHPIA 2.0 is designed to be adaptable to different home-based application scenarios, including human activity recognition, coaching systems, and occupancy detection and counting. The SHPIA platform is open source and freely available to the scientific community, fostering collaboration and innovation.

Blockchain based secure Ownership Transfer Protocol for smart objects in the Internet of Things

Secure digital ownership transfer is critical for smart objects within the Internet of Things (IoT) ecosystem. This paper presents the Ownership Transfer Protocol (OTP), which leverages Physically Unclonable Function (PUF) and blockchain technology to ensure the secure transfer of ownership for smart objects in the IoT. The proposed protocol can securely track and trace smart objects during their movement in the IoT supply chain. Unlike the traditional Ownership Transfer (OT) architecture, the proposed architecture does not require a Trusted Third Party (TTP) and can support Partial Ownership Transfer (POT). The innovative use of immutable blockchain architecture enabled the proposed protocol to effectively support distributed environments and authenticate both the device and involved parties. The proposed protocol is evaluated for its robustness against common attacks outlined in this paper and implemented using the Ethereum blockchain. The testbed results on Ethereum confirm the optimal gas consumption of the proposed model. Furthermore, utilizing the security claim verification tool, Scyther, the experiment validates the security claim regarding the communication between the parties involved in the proposed protocol’s OT process.

Web Platform for Interconnecting Body Sensors and Improving Health Care

The Internet of Things (IoT) is a paradigm in which smart objects actively collaborate among them and with other physical and virtual objects available in the Web in order to perform high-level tasks for the benefit of end-users. In the e-health scenario, these communicating smart objects can be body sensors that enable a continuous real-time monitoring of vital signs of patients. Data produced by such sensors can be used for several purposes and by different actors, such as doctors, patients, relatives, and health care centers, in order to provide remote assistance to users. However, major challenges arise mainly in terms of the interoperabil- ity among several heterogeneous devices from a variety of manufacturers. In this context, we introduce Eco Health (Ecosystem of Health Care Devices), a Web middleware platform for connecting doctors and patients using attached body sensors, thus aiming to provide improved health monitoring and diagnosis for patients. This platform is able to integrate information obtained from heterogeneous sensors in order to provide mechanisms to monitor, process, visualize, store, and send notifications regarding patients’ conditions and vital signs at real-time by using Internet standards. In this paper, we present blueprints of our proposal to Eco Health and its logical architecture and implementation, as well as an e-health motivational scenario where such a platform would be useful.

A Fully Privacy-Preserving Solution for Anomaly Detection in IoT using Federated Learning and Homomorphic Encryption

AbstractAnomaly detection for the Internet of Things (IoT) is a very important topic in the context of cyber-security. Indeed, as the pervasiveness of this technology is increasing, so is the number of threats and attacks targeting smart objects and their interactions. Behavioral fingerprinting has gained attention from researchers in this domain as it represents a novel strategy to model object interactions and assess their correctness and honesty. Still, there exist challenges in terms of the performance of such AI-based solutions. The main reasons can be alleged to scalability, privacy, and limitations on adopted Machine Learning algorithms. Indeed, in classical distributed fingerprinting approaches, an object models the behavior of a target contact by exploiting only the information coming from the direct interaction with it, which represents a very limited view of the target because it does not consider services and messages exchanged with other neighbors. On the other hand, building a global model of a target node behavior leveraging the information coming from the interactions with its neighbors, may lead to critical privacy concerns. To face this issue, the strategy proposed in this paper exploits Federated Learning to compute a global behavioral fingerprinting model for a target object, by analyzing its interactions with different peers in the network. Our solution allows the training of such models in a distributed way by relying also on a secure delegation strategy to involve less capable nodes in IoT. Moreover, through homomorphic encryption and Blockchain technology, our approach guarantees the privacy of both the target object and the different workers, as well as the robustness of the strategy in the presence of attacks. All these features lead to a secure fully privacy-preserving solution whose robustness, correctness, and performance are evaluated in this paper using a detailed security analysis and an extensive experimental campaign. Finally, the performance of our model is very satisfactory, as it consistently discriminates between normal and anomalous behaviors across all evaluated test sets, achieving an average accuracy value of 0.85.