IPTV Research Articles

An observing campaign to search for meteoroids of Bennu at Earth

An observing campaign was conducted in the Southern Hemisphere using low-light video camera triangulation to measure the trajectories and orbits of meteoroids with a possible origin at asteroid Bennu. New CAMS (Camera for Allsky Meteor Surveillance) video camera networks were established in Australia, Chile, and Namibia, and networks in New Zealand and South Africa were expanded. During observing periods in September 2019, 2020, and 2021, we measured 7672, 4936, and 5890 orbits, respectively. Based on the non-detection of predicted meteoroid trail encounters, Bennu's meteoroid production rate was <1.5 kg/s during 1500–1800 CE. Indeed, the current production rate is many orders of magnitude lower. Bennu may have an associated annual meteoroid stream of much older ejecta at a particle flux density of ≤1.3 × 10−6 km−2 h−1, based on seven Bennu-like orbits detected during the first three years of observations.

Effect of camera network configuration on the accuracy of digitization of architectural objects

Computer vision techniques enabled a significant boom in the use of photogrammetry in recent years, especially in the digitization of cultural heritage. Thanks to Structure from Motion, it is possible to automatically reconstruct the 3D geometry of even complicated objects. However, each technology has its strengths and weaknesses, so even SfM can often lead to unreliable results, which are usually caused by an inappropriate camera network configuration or an unsuitable camera calibration model. As part of this study, we analyzed the impact of different imaging configurations on the digitization of architectural objects, and in particular the possibility of automated photogrammetric linking of the exterior and interior through door and window openings. Photogrammetric point clouds were compared to scans from terrestrial laser scanning. The results indicate the possibility of reducing financial costs in the future if a laser scanner is not available.

Pose-driven attention-guided image generation for person re-Identification

Person re-identification (re-ID) concerns the matching of subject images across different camera views in a multi camera surveillance system. One of the major challenges in person re-ID is pose variations across the camera network, which significantly affects the appearance of a person. Existing development data lack adequate pose variations to carry out effective training of person re-ID systems. To solve this issue, in this paper we propose an end-to-end pose-driven attention-guided generative adversarial network, to generate multiple poses of a person. We propose to attentively learn and transfer the subject pose through an attention mechanism. A semantic-consistency loss is proposed to preserve the semantic information of the person during pose transfer. To ensure fine image details are realistic after pose translation, an appearance discriminator is used while a pose discriminator is used to ensure the pose of the transferred images will exactly be the same as the target pose. We show that by incorporating the proposed approach in a person re-identification framework, realistic pose transferred images and state-of-the-art re-identification results can be achieved.

UAV-BASED SURVEY: THE CASE OF COLONIA VARESE IN MILANO MARITTIMA

Abstract. The Maritime Colonies are an important part of Italy's cultural heritage buildings. The research focuses attention on a survey with an Unmanned Aerial Vehicles (UAV), using photogrammetry to determine features such as dimensions, object shape, volumes, areas, distances, elevations, and set up the reconstruction of a digital model of the Varese Colony, located in Milano Marittima, in the province of Ravenna. To digitize this structure, it was necessary to use a camera network capable of creating nadir and oblique images with different baselines. The proposed method has proven convincing in real-world application because of its workability, easy implementation, area coverage, and can be used in a wide range of applications on artifacts of many different morphologies.

Editorial to the special issue

This special issue of the International Journal of Technology Policy and Law considers recent developments in the reconfiguration of communication regulation to account for the impact of media convergence. It is readily apparent that media worldwide are going through a series of transformations, associated with the rise of the internet, user-created content and social media. The papers in the collection draw upon legal and policy developments in Australia, the European union and South Korea, and consider such issues as public participation in media policy and regulation, civic media governance for online platforms, the future copyright laws, the roles and responsibilities of internet intermediaries, and regulatory frameworks for internet protocol television (IPTV).

Social distance and face mask detector system exploiting transfer learning

&lt;p class="IJASEITAbtract"&gt;&lt;span lang="EN-US"&gt;As time advances, the use of deep learning-based object detection algorithms has also evolved leading to developments of new human-computer interactions, facilitating an exploration of various domains. Considering the automated process of detection, systems suitable for detecting violations are developed. One such applications is the social distancing and face mask detectors to control air-borne diseases. The objective of this research is to deploy transfer learning on object detection models for spotting violations in face masks and physical distance rules in real-time. The common drawbacks of existing models are low accuracy and inability to detect in real-time. The MobileNetV2 object detection model and YOLOv3 model with Euclidean distance measure have been used for detection of face mask and physical distancing. A proactive transfer learning approach is used to perform the functionality of face mask classification on the patterns obtained from the social distance detector model. On implementing the application on various surveillance footage, it was observed that the system could classify masked and unmasked faces and if social distancing was maintained or not with accuracies 99% and 94% respectively. The models exhibited high accuracy on testing and the system can be infused with the existing internet protocol (IP) cameras or surveillance systems for real-time surveillance of face masks and physical distancing rules effectively.&lt;/span&gt;&lt;/p&gt;

An IoT-fuzzy based password checker system for wireless video surveillance system

Wireless video surveillance systems (WVSS) are deployed in large environments for use in strategic places such as town centers, public streets, and airports and play an essential role in protecting critical infrastructure. However, WVSSs are vulnerable to unauthorized access due to weak login credentials, which leads to their exploitation to launch cyberattacks on other systems, such as distributed denial-of-service attacks. Hence, it is essential to secure these systems from unauthorized access. This paper proposes the Mamdani fuzzy inference system (FIS)-based password checker algorithm to estimate the password strength ratio (PSR) of internet protocol (IP) cameras and internet of things (IoT) devices. This algorithm composes three stages, the password extraction stage, which evaluates the input parameters of FIS from the real-time streaming protocol (RTSP) protocol using a counter of password characters. Then, the processing stage uses Mamdani FIS to optimize the input parameters to calculate the PSR. Finally, the alarm stage will notify the system administrator about weak IoT nodes. Unlike the existing approaches, this algorithm improves detection accuracy by informing the system administrator about threatened nodes. Extensive experiments are carried out to determine the efficiency of the proposed algorithm. The results confirm the efficiency of the proposed algorithm with high accuracy, which outperforms existing schemes.

Detecting and tracking multiple mobile radioactive sources by data fusion of a surveillance camera and a sodium iodide (NaI) detector.

Nowadays, the use of radioactive materials has increased due to recent progress in nuclear technology, but concerns about the missing or smuggling of radioactive materials have persisted. Nuclear threats such as terrorist attacks or the malicious use of radioactive materials out of regulatory control (MORC) are always severe problems that can result in adverse environmental, health, economic, and security effects. Applying new technologies to monitor and track MORC will prevent radioactive materials from being illegally transported across borders or from one place to another. In this research, we have constructed a detection system to automatically and remotely localize multiple mobile gamma-emitting radiation sources among other objects by combining an IP camera and a sodium iodide detector. An algorithm for the detection system has been developed to identify the objects' paths from camera data and correlate radiation data with the paths to detect contaminated objects. We evaluated the system using two weak radioactive sources (Co-60, Cs-137), which were hidden on moving objects, and succeeded in finding harmful targets. The results have also shown that by increasing the number of contaminated sources from one to two, the detection accuracy decreases by about 2.5 times. However, by doubling the speed of mobile targets, the detection accuracy improved by about 30%. Detecting and tracking MORC with a single detection system is limited to small regions. By equipping many surveillance cameras in a city with relatively inexpensive radioactive detectors, a network sensing system is established to find radioactive hotspots in a smart city.

Camera Image Processing on ESP32 Microcontroller with Help of Convolutional Neural Network

This paper analyzes a common ESP32 microcontroller with a built-in camera for image classification tasks using a convolutional neural network. ESP32 is commonly used in IoT devices to read data and control sensors, so its computing power is not significant, which has a positive effect on the cost of the device. The prevalence of ultra-low power embedded devices such as ESP32 will allow the widespread use of artificial intelligence built-in IoT devices. The duration of photographing and photo processing is obtained in the paper, as this can be a bottleneck of the microcontroller, especially together with machine learning algorithms. Deployed convolutional neural network, pre-trained on another device, MobileNet architecture on microcontroller and proved that ESP32 capacity is sufficient for simultaneous operation of both the camera and convolutional neural network.

Single-station meteor detection filtering using machine learning on MOROI data

ABSTRACT Nowadays, extensive data are collected in an automated regime. Combining this, with the increase in accessible computational power, led to large-scale implementations of machine learning (ML). This is also the case of meteor science, where object detection often requires tracking of a moving light source between frames, and the number of false positives can be up to an order of magnitude higher than true meteoric phenomena. While spatiotemporal coincidence of events recorded by close, multiple cameras can eliminate most of the false positives, single-station detections in some camera networks are currently discarded. In this paper, we explore a set of ML models aiming to find an optimal method for re-analysis of this single-station observations, in order to identify and extract real meteors. A set of 15 ML models were trained on features extracted from the meteor movement. Upon testing, we found a top accuracy score of 98,2 per cent, and a recall (i.e. percentage of meteors correctly classified) score of 96 per cent for the best performing models. When combined with the spatiotemporal coincidence of the detection, the recall increases to 99.92 per cent. These 15 ML techniques were selected according to their ability classify tabular data, hence the bundle can be applied to other studies. The same goes for the computed features, which are independent on the camera configuration, thus, the process can be scaled and applied to other networks. These methods are to be implemented to re-analyze the events recorded by the larger, FRIPON network.

V2ReID: Vision-Outlooker-Based Vehicle Re-Identification.

With the increase of large camera networks around us, it is becoming more difficult to manually identify vehicles. Computer vision enables us to automate this task. More specifically, vehicle re-identification (ReID) aims to identify cars in a camera network with non-overlapping views. Images captured of vehicles can undergo intense variations of appearance due to illumination, pose, or viewpoint. Furthermore, due to small inter-class similarities and large intra-class differences, feature learning is often enhanced with non-visual cues, such as the topology of camera networks and temporal information. These are, however, not always available or can be resource intensive for the model. Following the success of Transformer baselines in ReID, we propose for the first time an outlook-attention-based vehicle ReID framework using the Vision Outlooker as its backbone, which is able to encode finer-level features. We show that, without embedding any additional side information and using only the visual cues, we can achieve an 80.31% mAP and 97.13% R-1 on the VeRi-776 dataset. Besides documenting our research, this paper also aims to provide a comprehensive walkthrough of vehicle ReID. We aim to provide a starting point for individuals and organisations, as it is difficult to navigate through the myriad of complex research in this field.

Proxy methods for detection of inhalation exposure in simulated office environments

BackgroundModern health concerns related to air pollutant exposure in buildings have been exacerbated owing to several factors. Methods for assessing inhalation exposures indoors have been restricted to stationary air pollution measurements, typically assuming steady-state conditions.ObjectiveWe aimed to examine the feasibility of several proxy methods for estimating inhalation exposure to CO2, PM2.5, and PM10 in simulated office environments.MethodsIn a controlled climate chamber mimicking four different office setups, human participants performed a set of scripted sitting and standing office activities. Three proxy sensing techniques were examined: stationary indoor air quality (IAQ) monitoring, individual monitoring of physiological status by wearable wristband, human presence detection by Passive Infrared (PIR) sensors. A ground-truth of occupancy was obtained from video recordings of network cameras. The results were compared with the concurrent IAQ measurements in the breathing zone of a reference participant by means of multiple linear regression (MLR) analysis with a combination of different input parameters.ResultsSegregating data onto sitting and standing activities could lead to improved accuracy of exposure estimation model for CO2 and PM by 9–60% during sitting activities, relative to combined activities. Stationary PM2.5 and PM10 monitors positioned at the ceiling-mounted ventilation exhaust in vicinity of the seated reference participant accurately estimated inhalation exposure (adjusted R² = 0.91 and R² = 0.87). Measurement at the front edge of the desk near abdomen showed a moderate accuracy (adjusted R² = 0.58) in estimating exposure to CO2. Combining different sensing techniques improved the CO2 exposure detection by twofold, whereas the improvement for PM exposure detection was small (~10%).SignificanceThis study contributes to broadening the knowledge of proxy methods for personal exposure estimation under dynamic occupancy profiles. The study recommendations on optimal monitor combination and placement could help stakeholders better understand spatial air pollutant gradients indoors which can ultimately improve control of IAQ.

Data on 824 fireballs observed by the digital cameras of the European Fireball Network in 2017–2018

A catalog of 824 fireballs (bright meteors), observed by a dedicated network of all-sky digital photographic cameras in central Europe in the years 2017–2018 is presented. The status of the European Fireball Network, established in 1963, is described. The cameras collect digital images of meteors brighter than an absolute magnitude of about −2 and radiometric light curves with a high temporal resolution of those brighter than a magnitude ≈−4. All meteoroids larger than 5 g, corresponding to sizes of about 2 cm, are detected regardless of their entry velocity. High-velocity meteoroids are detected down to masses of about 0.1 g. The largest observed meteoroid in the reported period 2017–2018 had a mass of about 100 kg and a size of about 40 cm. The methods of data analysis are explained and all catalog entries are described in detail. The provided data include the fireball date and time, atmospheric trajectory and velocity, the radiant in various coordinate systems, heliocentric orbital elements, maximum brightness, radiated energy, initial and terminal masses, maximum encountered dynamic pressure, physical classification, and possible shower membership. Basic information on the fireball spectrum is available for some bright fireballs (apparent magnitude &lt;−7). A simple statistical evaluation of the whole sample is provided. The scientific analysis is presented in an accompanying paper.

Seamless Multiple-Target Tracking Method Across Overlapped Multiple Camera Views Using High-Speed Image Capture

Multiple-target tracking across multiple camera views is required for various practical tracking and surveillance applications. This study proposes a seamless tracking method based on label transfer using high-speed image capture. The proposed method does not require camera calibration to estimate the extrinsic parameters, and can be applied with simple line calibration. Therefore, it is effective for large-scale camera networks. In addition, owing to the features of the high-speed image capture and self-window, our method ensures seamless tracking of objects across multiple camera views. The feasibility of the method is proven through an experiment using a high-speed camera network system.

Advanced technologies for enhanced construction safety management: investigating Malaysian perspectives

The global construction industry continues to struggle with safety issues. Although the existing literature suggests that the application of innovative solutions can reduce workplace accidents and improve work safety, the adoption of safety technologies is still limited, particularly in the developing world. This paper aims to investigate the capacity of advanced technologies to enhance safety management and those suited to the Malaysian context. Following a synthesis of the literature, a questionnaire survey was developed comprising 9 potential applications and 15 types of safety technologies. The feedback from 150 local construction professionals was analysed. The leading benefits are related to advancing safety planning, building safety awareness, delivering effective safety training, improving safety inspections and optimising hazard identification. The most salient technologies for safety management are building information modelling (BIM), camera network systems, mobile devices on site, internet of things (IoT) and digital signage. The findings will provide industry practitioners with an increased understanding of the potential of integrating technological solutions into the various tasks associated with construction safety management. This study highlighted the capabilities of advanced technologies that will revolutionise the future of the construction safety landscape and the need for new technology-based policies, procedures, and practices.

Towards an Automated Approach for Monitoring Tree Phenology Using Vehicle Dashcams in Urban Environments.

Trees in urban environments hold significant value in providing ecosystem services, which will become increasingly important as urban populations grow. Tree phenology is highly sensitive to climatic variation, and resultant phenological shifts have significant impact on ecosystem function. Data on urban tree phenology is important to collect. Typical remote methods to monitor tree phenological transitions, such as satellite remote sensing and fixed digital camera networks, are limited by financial costs and coarse resolutions, both spatially and temporally and thus there exists a data gap in urban settings. Here, we report on a pilot study to evaluate the potential to estimate phenological metrics from imagery acquired with a conventional dashcam fitted to a car. Dashcam images were acquired daily in spring 2020, March to May, for a 2000 m stretch of road in Melksham, UK. This pilot study indicates that time series imagery of urban trees, from which meaningful phenological data can be extracted, is obtainable from a car-mounted dashcam. The method based on the YOLOv3 deep learning algorithm demonstrated suitability for automating stages of processing towards deriving a greenness metric from which the date of tree green-up was calculated. These dates of green-up are similar to those obtained by visual analyses, with a maximum of a 4-day difference; and differences in green-up between trees (species-dependent) were evident. Further work is required to fully automate such an approach for other remote sensing capture methods, and to scale-up through authoritative and citizen science agencies.

Low Power Environmental Image Sensors for Remote Photogrammetry.

This paper aims to prove the feasibility of a 4D monitoring solution (3D modeling and temporal monitoring) for the sandbar and to characterize the species’ role in the landscape. The developed solution allows studying the interaction between the river dynamics and vegetation using a network of low resolution and low power sensors. The issues addressed concern the feasibility of implementing a photogrammetry solution using low-resolution sensors as well as the choice of the appropriate sensor and its testing according to different configurations (image capture and storage on the sensor and/or image transmission to a centralization node) and also the detailed analysis of the different phases of the process (camera initialization, image capture, network transmission and selection of the most appropriate standby mode). We reveal that the tiny, low-cost board (ESP32-Cam) can perform a 3D reconstruction and propose using the camera’s UXGA (1600, 1200) resolution because of the quality rendering and energy consumption. A multi-node scenario based on a combined Wi-Fi and GSM relay is proposed in the study showing several years of autonomy for the system. Finally, to illustrate the energy cost of the module, we have defined a study process, where we have identified and quantified one by one the different phases of operation of the card for better energy optimization (setup, camera configuration, shooting, saving on SD card, or sending by Wi-Fi). The device is now operational for deployment on the Allier River (France).

Camera and LiDAR-based point painted voxel region-based convolutional neural network for robust 3D object detection

Most of the three-dimensional (3D) object detection methods based on LiDAR point cloud data achieve relatively high performance in general cases. However, when the LiDAR points have noise or some corruptions, the detection performance can be severely affected. We propose a 3D object detection method that combines point cloud information with two-dimensional (2D) semantic segmentation information to enhance the feature representation for difficult cases, such as sparse, noisy, and partially absent data. Motivated by the Pointpainting techniques, we designed an early-stage fusion method based on a Voxel region-based convolutional neural network (R-CNN) architecture. The 2D semantic segmentation scores obtained by the Pointpainting techniques are appended to the raw point cloud data. The voxel-based features and 2D semantic information improve the performance in detecting instances when the point cloud is corrupted. In addition, we also designed a multiscale hierarchical region of interest pooling strategy that reduced the computational cost of Voxel R-CNN by at least 43%. Our method shows competitive results with the state-of-the-art methods on the standard KITTI dataset. In addition, three corrupted KITTI datasets, KITTI sparse (KITTI-S), KITTI jittering (KITTI-J), and KITTI dropout (KITTI-D), were used for robustness testing. With the noisy LiDAR points, our proposed point painted Voxel R-CNN achieved superior detection performance over that of the baseline Voxel R-CNN for the moderate case, with a notable improvement of 11.13% in average precision (AP) on the 3D object detection and 14.3% in AP on the bird’s eye view object detection.

Novel lightweight video encryption method based on ChaCha20 stream cipher and hybrid chaotic map

&lt;span lang="EN-US"&gt;In the recent years, an increasing demand for securing visual resource-constrained devices become a challenging problem due to the characteristics of these devices. Visual resource-constrained devices are suffered from limited storage space and lower power for computation such as wireless sensors, internet protocol (IP) camera and smart cards. Consequently, to support and preserve the video privacy in video surveillance system, lightweight security methods are required instead of the existing traditional encryption methods. In this paper, a new light weight stream cipher method is presented and investigated for video encryption based on hybrid chaotic map and ChaCha20 algorithm. Two chaotic maps are employed for keys generation process in order to achieve permutation and encryption tasks, respectively. The frames sequences are encrypted-decrypted based on symmetric scheme with assist of ChaCha20 algorithm. The proposed lightweight stream cipher method has been tested on several video samples to confirm suitability and validation in term of encryption–decryption procedures. The performance evaluation metrics include visual test, histogram analysis, information entropy, correlation analysis and differential analysis. From the experimental results, the proposed lightweight encryption method exhibited a higher security with lower computation time compared with state-of-the-art encryption methods.&lt;/span&gt;

Implementasi Metode Perbandingan Eksponensial (MPE) Pada Sistem Pendukung Keputusuan Pemilihan Internet Protocol Camera

The development of video surveillance has given rise to various types of surveillance cameras, one of which is the Internet Protocol Camera (IP Camera). The number of IP Camera brands in the market, makes people who want to buy IP Cameras have to find their own information about the specifications and capabilities of the IP Camera to be purchased. It takes time and effort to choose an IP Camera, because you have to learn one by one which IP Camera to buy. This study aims to build a decision support system for choosing an IP Camera with a website-based Exponential Comparison Method (MPE) to make it easier to determine the right IP Camera. MPE can sort the priority of decision alternatives on existing criteria and is able to distinguish the value of each alternative in contrast. Based on the case study, the best alternative is Xiaomi Mi 360 with a value of 386, followed by Yi Home Camera 3 getting a value of 369, Ezviz C6N getting a value of 350, Imilab EC4 getting a value of 343 and Cleverdog Egg Cam getting a value of 110. The results of the MPE calculation generated by the system shows the same value as the manual calculation, then the MPE calculation on the system is declared valid. In addition, the test results with black-box testing show that the system can run well.