City Surveillance Research Articles

Editor’s Note: In Defense of Freedoms: Decentralized Networks as a Shield Against AI-Powered State Surveillance in Smart Cities

Associate Editor Ahmed Alrawi discusses the advancement of AI-powered state surveillance in smart cities and how they have sparked concerns. Decentralized systems may present a solution. Dispersed dat and systems make it more difficult for government security agencies and urban administrators to conduct widespread surveillance but also empower individuals with greater control over their personal data.

Two-timescale joint service caching and resource allocation for task offloading with edge–cloud cooperation

Task offloading with edge–cloud cooperation has emerged as a pivotal solution for meeting the intricate array of application coupled with dynamically evolving business demand in 6G business scenarios, such as traffic sensing, environmental monitoring, and video surveillance in smart cities. Nonetheless, effectively leveraging heterogeneous edge–cloud network resources for effective task offloading presents substantial challenges. Additionally, the inherent differences in system decision cycles escalate the complexity of the task offloading problem to a new dimension. In this study, we delve into a two-timescale joint service caching and resource allocation optimization for task offloading within edge–cloud cooperation aiming to maximize long-term network performance while adhering to energy constraints. We propose a novel edge–cloud cooperation task offloading scheme that supports both edge–cloud and edge–edge cooperation to effectively balance the edge–cloud and edge–edge loads, promoting the efficient co-utilization of all edge–cloud system resources. Furthermore, we devise an online two-timescale Lyapunov-based joint optimization framework for service caching, task offloading, and computing resource allocation. Our two-timescale decision-making framework can flexibly accommodate the inherent differences in the sensitive decision optimization periods, thereby mitigating the degradation of task offloading performance caused by frequent service caching updates. Finally, theoretical analysis confirms that our proposed algorithm can converge to an approximate optimal solution in polynomial time, and the superiority of our scheme is validated by extensive simulation experiments.

An Adversarial Machine Learning Based Approach for Privacy Preserving Face Recognition in Distributed Smart City Surveillance

Smart cities rely heavily on surveillance cameras for urban management and security. However, the extensive use of these cameras also raises significant concerns regarding data privacy. Unauthorized access to facial data captured by these cameras and the potential for misuse of this data poses serious threats to individuals’ privacy. Current privacy preservation solutions often compromise data usability with noise application-based approaches and vulnerable centralized data handling settings. To address these privacy challenges, we propose a novel approach that combines Adversarial Machine Learning (AML) with Federated Learning (FL). Our approach involves the use of a noise generator that perturbs surveillance data right from the source before they leave the surveillance cameras. By exclusively training the Federated Learning model on these perturbed samples, we ensure that sensitive biometric features are not shared with centralized servers. Instead, such data remains on local devices (e.g., cameras), thereby ensuring that data privacy is maintained. We performed a thorough real-world evaluation of the proposed method and achieved an accuracy of around 99.95% in standard machine learning settings. In distributed settings, we achieved an accuracy of around 96.24% using federated learning, demonstrating the practicality and effectiveness of the proposed solution.11The code is available at: https://github.com/farah-wahida/Privacy-Preserving-Face-Recognition-in-Distributed-Smart-City-Surveillance.

Efficiently adapting large pre-trained models for real-time violence recognition in smart city surveillance

Efficiently adapting large pre-trained models for real-time violence recognition in smart city surveillance

Surveillance for rhabdomyolysis after the consumption of crayfish in Wuhan, China, 2016-2022.

To analyze the epidemiological characteristics and etiology of crayfish-related rhabdomyolysis. Cases of crayfish-related rhabdomyolysis in Wuhan were monitored, and professional training of city's surveillance personnel was conducted. Unified questionnaires were used to collect data. The first case of crayfish-related rhabdomyolysis occurred on July 12, 2016. Subsequently, 423 patients were reported over the next 7 years, with muscle pain, weakness, and chest distress as main symptoms. In total, 64.54% (273/423) of patients were females, and young adults (aged 20-49 years) account for 86.22% (363/423) of patients. The primary clinical presentations were muscle pain, muscle weakness, and chest discomfort. The median incubation time was 6 h. And the number of cases may be related to water levels in Yangzi river. Laboratory tests revealed elevated creatine kinase and myoglobin levels. In total, 95.16% (236/248) of patients had consumed crayfish tail shrimp and 91.53% (227/248) had consumed crayfish liver and pancreas (Female crayfish also contain ovaries). Only 25.00% (62/248) of patients had a history of alcohol consumption. On average, 227 patients consumed 15 (3-50) crayfish, of whom 84.14% (191/227) consumed more than 10 crayfish. All patients had a favorable prognosis. Crayfish-related rhabdomyolysis is a kind of a case or cluster of patients present with severe myalgia or weakness of unknown etiology and mechanism disease in Wuhan, China, 2016-2022. Excessive consumption of crayfish may be a risk factor for the disease. The relationship between the specific parts of crayfish consumed and the onset of the disease is unclear, suggesting further research is needed to identify the relevant risk factors for the disease.

Smart surveillance using IoT: a review

In today’s modern society, video surveillance is a growing trend and it can revolutionize many aspects of technology, especially in future smart cities that will transform traditional surveillance systems into intelligent and interconnected networks. It may be difficult for even well-trained employees to process and respond immediately to monitored data. Moreover, IoT-enabled surveillance systems overcome the challenges and flaws of conventional passive monitoring techniques by offering real-time surveillance and automated notifications for suspicious activity, intrusions, or anomalies. Therefore, the objective of this article is to provide an Internet of Things (IoT)-based smart surveillance system that may successfully connect the ecosystem, resulting in enhanced monitoring for smart city services. The primary goal of this study is to provide a comprehensive review of several IoT-based surveillance techniques used in various smart city applications. The categorization of tasks for each section is as follows: to present the historical context overview; to examine the significance of IoT and its application in smart cities; to present a standardized architecture for IoT-based smart city surveillance; to highlight an authoritative and thorough review of current IoT-based smart surveillance systems; and to identify various research issues. The methods used are: statistical graph or chart approach, schematic, and timeline diagram. Conclusions. This article outlines numerous research challenges for future video surveillance that may be addressed by researchers. In summary, this comprehensive review provides a valuable and streamlined resource for future researchers exploring smart city surveillance through the IoT.

Continuous Authentication for Consumer Electronics in Smart City Surveillance

Continuous Authentication for Consumer Electronics in Smart City Surveillance

AI-based outdoor moving object detection for smart city surveillance

<abstract> <p>One essential component of the futuristic way of living in "smart cities" is the installation of surveillance cameras. There are a wide variety of applications for surveillance cameras, including but not limited to: investigating and preventing crimes, identifying sick individuals (coronavirus), locating missing persons, and many more. In this research, we provided a system for smart city outdoor item recognition using visual data collected by security cameras. The object identification model used by the proposed outdoor system was an enhanced version of RetinaNet. A state of the art object identification model, RetinaNet boasts lightning-fast processing and pinpoint accuracy. Its primary purpose was to rectify the focal loss-based training dataset's inherent class imbalance. To make the RetinaNet better at identifying tiny objects, we increased its receptive field with custom-made convolution blocks. In addition, we adjusted the number of anchors by decreasing their scale and increasing their ratio. Using a mix of open-source datasets including BDD100K, MS COCO, and Pascal Vocab, the suggested outdoor object identification system was trained and tested. While maintaining real-time operation, the suggested system's performance has been markedly enhanced in terms of accuracy.</p> </abstract>

Crowd Emotion Prediction for Human-Vehicle Interaction Through Modified Transfer Learning and Fuzzy Logic Ranking

In metropolitan environments, unmanned aerial vehicles (UAVs) equipped with video surveillance equipment can monitor crowd behavior and maintain public safety. In high-traffic areas where humans are more likely to make mistakes, a smart city needs modern technology to forecast the behavior of its residents. In order to improve citywide traffic flow, urban transportation systems (UTSs) monitor and learn how people behave in crowds. Using UAVs for video surveillance in smart cities, our research describes a unique way to assess crowd condition, which expands the scope of human-vehicle interactions. Moreover, we use fuzzy logic ranking to improve the system’s ability to detect anomalies in crowds. In order to improve decision-making, a novel deep transfer learning (DTL) technique is applied to the UAV’s received frames. A 98.5% accuracy rate, satisfactory performance, and robustness to population behavior are all characteristics of the proposed integrated model. In UTSs and urban areas, our novel intelligent system analyzes human behavior based on vehicle-human interactions. In areas with low and high traffic congestion, the modified ResNet (mResNet) architecture predicts the crowd’s condition based on fuzzy logic (FLA). Through decision-making based on accurate crowd conditions, best general paths can be selected using the algorithm.

Pre-exposure Prophylaxis Use History in People With Antiretroviral Resistance at HIV Diagnosis: Findings From New York City HIV Surveillance and Partner Services, 2015-2022.

Drug resistance may be acquired in people starting human immunodeficiency virus (HIV) preexposure prophylaxis (PrEP) during undiagnosed infection. Population-based estimates of PrEP-related resistance are lacking. We used New York City surveillance and partner services data to measure the effect of PrEP use (tenofovir disoproxil fumarate/tenofivir alafenamide fumarate with emtricitabine) history on the baseline prevalence of M184I/V mutations in people with HIV diagnosed in 2015-2022. PrEP use was categorized as "recent" (defined as PrEP stopped ≤90 days before diagnosis), "past" (PrEP stopped >90 days before diagnosis), or "no known use." Resistance-associated mutations were determined using the Stanford algorithm. We used log binomial regression to generate the adjusted relative risk (aRR) of M184I/V by PrEP use history in people with or without acute HIV infection (AHI). Of 4246 people with newly diagnosed HIV and a genotype obtained within ≤30 days of diagnosis, 560 (13%) had AHI; 136 (3%) reported recent and 124 (35%) past PrEP use; and 98 (2%) harbored M184I/V. In people with AHI, recent PrEP use was associated with a 6 times greater risk of M184I/V than no known use (aRR, 5.86 [95% confidence interval, 2.49-13.77]). Among people without AHI, the risk of M184I/V in recent users was 7 times that in people with no known use (aRR, 7.26 [95% confidence interval, 3.98-13.24]), and in past users, it was 4 times that in those with no known use (4.46 [2.15-9.24]). PrEP use was strongly associated with baseline M184I/V in New York City, regardless of AHI status. Ordering a nucleic acid test when indicated after assessment of exposure, antiretroviral history, and AHI symptoms can decrease PrEP initiation in people with undetected infection.

From subjects to subversives: Chinese migrants and the evolution of the French colonial surveillance regime in Sài Gòn-Chợ Lớn, 1874-1930

ABSTRACT In 1874, the French colonial state introduced an extensive regime of ethnic surveillance in the port cities of Sài Gòn-Chợ Lớn. This pioneering system in Vietnam relied on an innovative anthropometric science to target Chinese migrants, whose mobile networks posed ongoing concerns for the imperial authority. This article explores the strategies employed by Chinese migrants to navigate the enforced ethnic classifications and regulations imposed by the colonial administration, exploiting the uncertain boundaries between subjects, foreigners, and citizens. It argues that Chinese subversions emerged through a dynamic interplay with the evolving French bureaucratic practices, shaped by the coexistence of two overlapping imperial systems: a decentralized Chinese empire and a French colonial state aiming to consolidate its rule in Cochinchina. The article reveals that French colonial surveillance was not a one-sided, rigid process of panoptic imperial dominance, but rather a complex landscape characterized by negotiated coexistence, mutual collaborations, and acts of resistance.

A Novel Unsupervised Video Anomaly Detection Framework Based on Optical Flow Reconstruction and Erased Frame Prediction.

Reconstruction-based and prediction-based approaches are widely used for video anomaly detection (VAD) in smart city surveillance applications. However, neither of these approaches can effectively utilize the rich contextual information that exists in videos, which makes it difficult to accurately perceive anomalous activities. In this paper, we exploit the idea of a training model based on the "Cloze Test" strategy in natural language processing (NLP) and introduce a novel unsupervised learning framework to encode both motion and appearance information at an object level. Specifically, to store the normal modes of video activity reconstructions, we first design an optical stream memory network with skip connections. Secondly, we build a space-time cube (STC) for use as the basic processing unit of the model and erase a patch in the STC to form the frame to be reconstructed. This enables a so-called "incomplete event (IE)" to be completed. On this basis, a conditional autoencoder is utilized to capture the high correspondence between optical flow and STC. The model predicts erased patches in IEs based on the context of the front and back frames. Finally, we employ a generating adversarial network (GAN)-based training method to improve the performance of VAD. By distinguishing the predicted erased optical flow and erased video frame, the anomaly detection results are shown to be more reliable with our proposed method which can help reconstruct the original video in IE. Comparative experiments conducted on the benchmark UCSD Ped2, CUHK Avenue, and ShanghaiTech datasets demonstrate AUROC scores reaching 97.7%, 89.7%, and 75.8%, respectively.

Citizen empowerment through smart surveillance: evidence from Indian smart cities

PurposeCitizens often perceive surveillance by government authorities as oppressive and, hence, demonstrate reluctance in value co-creation from such services. This study aims to investigate the challenges and benefits of citizen empowerment through technology-driven surveillance or “smart surveillance.”Design/methodology/approachGuided by Dynamic Capability theory, the authors conduct in-depth interviews with officers in-charge of surveillance in smart cities. Given the contemporary advancements, this approach allows a retrospective and real-time understanding of interviewees’ experiences with smart surveillance.FindingsThe authors develop five propositions for citizen empowerment through smart surveillance to summarize the findings of this study.Research limitations/implicationsThis study advances the relevance of Dynamic Capability in public administration.Practical implicationsSmart city authorities and policymakers may leverage the insights provided in this study to design appropriate policies for smart surveillance.Originality/valueThe authors find that factors such as digital technology and infrastructure, information management, skill divide and perceived return on investment may influence citizen empowerment through smart surveillance.

Aerial Surveillance in Cities: When UAVs Take Public Transportation Vehicles

This paper considers using unmanned aerial vehicles (UAVs) to survey important sites across a city. When the sites are relatively far from the UAVs’ depot, the UAVs may not be able to reach many of the sites. Suppose that a UAV can take public transportation vehicles (PTVs) like a passenger. Then, it may reach a site that is unreachable by flying only. Based on this UAV-PTV scheme, we investigate a task-UAV assignment problem, which assigns a set of surveillance tasks to UAVs. We formulate a mixed-integer linear programming (MILP) problem that minimizes the overall energy consumption of UAVs, subject to that every site is surveyed by a certain number of UAVs during a given time window, and all UAVs successfully return to the depot. Considering that this problem is NP-hard, we present two sub-optimal solutions. The first solution orders the surveillance tasks according to the starting times of their time windows. Then, starting from the earliest one, it assigns the tasks one by one to UAVs. The second solution breaks the tasks into small non-overlapping groups. It then assigns tasks to UAVs group by group. The former solution quickly addresses the assignment problem, but it lacks the overall management of UAV resources. The latter improves this by assigning a group of tasks simultaneously, and it can control the computation complexity by limiting the group size. The comparison with the brute force method shows that the proposed solutions can achieve competitive performance in a reasonable time. Note to Practitioners—Unmanned aerial vehicles (UAVs) have been widely used in surveillance missions. However, one challenge practitioners often meet is the limited flight duration. Commercial UAVs are in general powered by the onboard battery. Due to the restriction of payload, the battery capacity is constrained, which limits the UAVs’ operation time. In this paper, we present the approach exploiting public transportation vehicles (PTVs). In our design, a UAV can take a public transportation vehicles such as buses, trams and trains on the roof and transfer between vehicles when necessary. With this UAV-PTV collaboration scheme, we consider how to efficiently assign surveillance tasks to UAVs. Due to the NP-hardness of the considered problem, two suboptimal algorithms are presented.

Making Smart Things Strange Again: Using Walking as a Method for Studying Subjective Experiences of Smart City Surveillance

Smart cities are commonly seen as places that are defined by surveillance because of their reliance on vast amounts of digital data to improve urban management challenges. Although the infrastructures and technologies that enable smart city surveillance pervade multitudinous urban spaces and everyday practices, they are often “hiding in plain sight,” going unnoticed in the bustle of everyday life. Hence, fostering research settings where citizens can productively reflect on their everyday surveillance constitutes a major challenge for the interrelated projects of doing empirical research about subjective experiences of smart city surveillance and the inclusion of citizens in smart city discussions. Drawing on walking as a method, this study attempts to meet this challenge by developing and empirically testing a methodology of purposive “data walking.” Situating the research in Rotterdam, the Netherlands, participants are instructed to identify data points for public safety purposes on a short walk through the city and reflect on their experiences. Observations and experiences of smart city surveillance are documented with photos, text descriptions, and audio notes, which are shared in real-time with researchers and provide the basis for group reflections. These walks and reflections generate rich visual and textual data that yield insights into embodied and situated constructions of smart city surveillance as an object of subjective inquiry, experiences of visibility, considerations of agency and evaluations of public safety implications. The study considers these empirical results in conjunction with reflections on the methodology, contributing to further methodological explorations for including citizens in smart city discussions and surveillance subjectivity research.

A City Surveillance System for Social Health Inequalities: The Case of Barcelona.

In the past, health inequalities were not prioritised in the political agenda of Barcelona. The change of city government (2015) was an opportunity to develop a Surveillance System for Social Health Inequalities in the city, which is described in this article. The design of the Surveillance System formed part of the Joint Action for Health Equity in Europe (JAHEE), funded by the European Union. Various steps were considered by the experts to set up the System: define its objectives, target population, domains and indicators, and sources of information; perform data analysis; implement and disseminate the system; define the evaluation; and perform regular data updates. The System considers the following domains: social determinants of health, health-related with behaviours, use of healthcare, and health outcomes, and includes eight indicators. As axes of inequality, the experts chose sex, age, social class, country of origin, and geographical area. The Surveillance System for Social Health Inequalities is presented on a website including different types of figures. The methodology used to implement the Surveillance System can be used to implement similar systems in other urban areas around the world.

Vision transformer attention with multi-reservoir echo state network for anomaly recognition

Anomalous event recognition requires an instant response to reduce the loss of human life and property; however, existing automated systems show limited performance due to considerations related to the temporal domain of the videos and ignore the significant role of spatial information. Furthermore, although current surveillance systems can detect anomalous events, they require human intervention to recognise their nature and to select appropriate countermeasures, as there are no fully automatic surveillance techniques that can simultaneously detect and interpret anomalous events. Therefore, we present a framework called Vision Transformer Anomaly Recognition (ViT-ARN) that can detect and interpret anomalies in smart city surveillance videos. The framework consists of two stages: the first involves online anomaly detection, for which a customised, lightweight, one-class deep neural network is developed to detect anomalies in a surveillance environment, while in the second stage, the detected anomaly is further classified into the corresponding class. The size of our anomaly detection model is compressed using a filter pruning strategy based on a geometric median, with the aim of easy adaptability for resource-constrained devices. Anomaly classification is based on vision transformer features and is followed by a bottleneck attention mechanism to enhance the representation. The refined features are passed to a multi-reservoir echo state network for a detailed analysis of real-world anomalies such as vandalism and road accidents. A total of 858 and 1600 videos from two datasets are used to train the proposed model, and extensive experiments on the LAD-2000 and UCF-Crime datasets comprising 290 and 400 testing videos reveal that our framework can recognise anomalies more effectively, outperforming other state-of-the-art approaches with increases in accuracy of 10.14% and 3% on the LAD-2000 and UCF-Crime datasets, respectively.

Unsupervised Face Detection in the Dark.

Low-light face detection is challenging but critical for real-world applications, such as nighttime autonomous driving and city surveillance. Current face detection models rely on extensive annotations and lack generality and flexibility. In this paper, we explore how to learn face detectors without low-light annotations. Fully exploiting existing normal light data, we propose adapting face detectors from normal light to low light. This task is difficult because the gap between brightness and darkness is too large and complicated at the object level and pixel level. Accordingly, the performance of current low-light enhancement or adaptation methods is unsatisfactory. To solve this problem, we propose a joint High-Low Adaptation (HLA) framework. We design bidirectional low-level adaptation and multitask high-level adaptation. For low-level, we enhance the dark images and degrade the normal-light images, making both domains move toward each other. For high-level, we combine context-based and contrastive learning to comprehensively close the features on different domains. Experiments show that our HLA-Face v2 model obtains superior low-light face detection performance even without the use of low-light annotations. Moreover, our adaptation scheme can be extended to a wide range of applications, such as improving supervised learning and generic object detection. Project publicly available at: https://daooshee.github.io/HLA-Face-v2-Website/.

An efficient deep learning-assisted person re-identification solution for intelligent video surveillance in smart cities

An efficient deep learning-assisted person re-identification solution for intelligent video surveillance in smart cities

Comments on “A Secure and Lightweight Drones-Access Protocol for Smart City Surveillance”

Recently, Akram et al. (2021) proposed a drones-access protocol for monitoring urban security. Akram et al. claimed that their protocol is secure and can resist known security attacks. However, in this comment article, we demonstrate that their protocol is still vulnerable to drone capture attacks and stolen-verifier attacks. In addition, their protocol does not provide perfect forward secrecy. Finally, we present an improved protocol to manage the security weaknesses we found.