Camera Placement Research Articles

THE RISE OF ULTRAVIOLET ASTRONOMY IN FRANCE

France was one of the few countries able to launch rockets soon after the end of WWII. Some of these rockets were used to observe the Sun and other astronomical targets in the ultraviolet in the early 1960s. This program continued with the placement of French UV cameras and spectrographs in the French satellite D2B-Aura (1975), in two Soviet satellites and in NASA's Skylab (1973) and Spacelab (1983 and 1985). In parallel, a stratospheric balloon program was initiated around 1978, which has been considerably more successful than the rocket and satellite programs. We describe these activities and their scientific results, which culminated in an important French participation in the focal instrumentation of the Hubble Space Telescope and of several NASA and ESA ultraviolet satellites.

View-synthesized ‘re-calibration’ of an array projector for 3D measurement from an arbitrary monocular view

We previously presented a method on how to calibrate a high-speed Multi-Aperture Array Projector (MAAP) in order to obtain three-dimensional (3D) surface measurements using only a single camera instead of a stereo-camera configuration. The MAAP calibration procedure however can be time-consuming as it requires the camera to encode each aperture's illumination frustum by imaging the pattern incident on a depth-wise scanned test plane. This is as if recording discretized depth slices of the illumination frustums. This can be detrimental as the calibration is coupled to a single specific camera view. It would be laborious and time-consuming to repeat the entire MAAP calibration process for an alternate camera view. This study demonstrates a so-called MAAP ‘re-calibration’ solution that artificially synthesizes MAAP calibration images from an arbitrary camera view without direct imaging. This presents the opportunity to decouple the MAAP calibration from a single camera view and obtain 3D measurements from alternate camera placement setups without having to perform the entire MAAP calibration process again. This study shows that the synthesized MAAP calibration images resemble well the ground truth images. As a result, the proposed method has similar 3D measurement performance compared to when calibrating the MAAP with true images.

Between Journalist Authorship and User Agency: Exploring the Concept of Objectivity in VR Journalism

ABSTRACT Virtual Reality (VR) journalism has brought many changes to journalistic practice, calling facticity, accuracy, objectivity and autonomy into question. This study examines factors influencing objectivity and its significance in VR journalism from the perspective of VR content creators. The researcher conducted 27 semi-structured interviews with fact-based VR content creators. The findings showed the diverse understandings of objectivity among VR content creators and the growing acceptance of VR as a subjective experience. The 360° spherical view reduces the ability to frame the content. The different story-paths in interactive VR storytelling may add perspectives to the story. However, the VR user retains the ability to actively explore the virtual environment, deciding what to experience and what to dismiss. VR emphasises user-induced subjectivity in addition to the already existing journalist-induced subjectivity, in which the journalist chooses camera placement and provides a guided experience. Subjectivity is inescapable in VR and its impact is magnified, suggesting extra caution from VR content creators to protect users and affirming the need for the notion of pragmatic objectivity.

Real-Time Vehicle Orientation Classification and Viewpoint-Aware Vehicle Re-Identification

Vehicle re-identification (re-ID) is based on identity matching of vehicles across non-overlapping camera views. Recently, the research on vehicle re-ID attracts increased attention, mainly due to its prominent industrial applications, such as post-crime analysis, traffic flow analysis, and wide-area vehicle tracking. However, despite the increased interest, the problem remains to be challenging. One of the most significant difficulties of vehicle re-ID is the large viewpoint variations due to non-standardized camera placements. In this study, to improve re-ID robustness against viewpoint variations while preserving algorithm efficiency, we exploit the use of vehicle orientation information. First, we analyze and benchmark various deep learning architectures in terms of performance, memory use, and cost on applicability to orientation classification. Secondly, the extracted orientation information is utilized to improve the vehicle re-ID task. For this, we propose a viewpoint-aware multi-branch network that improves the vehicle re-ID performance without increasing the forward inference time. Third, we introduce a viewpoint-aware mini-batching approach which yields improved training and higher re-ID performance. The experiments show an increase of 4.0% mAP and 4.4% rank-1 score on the popular VeRi dataset with the proposed mini-batching strategy, and overall, an increase of 2.2% mAP and 3.8% rank-1 score compared to the ResNet-50 baseline.

Efficient Real-Time Human Detection Using Unmanned Aerial Vehicles Optical Imagery

ABSTRACT Unmanned Aerial Vehicles (UAVs) are promising technologies within many different application scenarios including human detection in search and rescue and surveillance use cases, which have received considerable attention worldwide. However, adverse conditions, such as varying altitude, overhead camera placement, changing illumination and moving platform, impose challenges for high-performance yet cost-efficient human detection. To overcome these challenges, we propose a novel combination of dilated convolutions with Path Aggregation Network (PAN) as a new deep neural network-based human detection algorithm in real time. Furthermore, we establish a comprehensive human detection dataset with varying backgrounds, illuminations, and contrast and train the proposed machine-learning model on the collected dataset. Our approach achieves both high precision (88.0% mean Average Precision (mAP)) and real time (67.0 Frames Per Second (FPS)) on a commercial off-the-shelf PC platform. In terms of accuracy, the result is comparable to the standard You Only Look Once v3 (YOLOv3). However, the speed is twice as that of the standard YOLOv3. YOLOv4 is slightly more accurate (89.8%) than our approach. However, it is slower (38.0 versus 67.0 FPS) and has more Billion Floating-Point Operations (BFLOPS). The proposed algorithm has also trained with the VisDrone2019 dataset and compared with seven studies using this dataset. The results have further validated the effectiveness of the proposed approach. Moreover, the algorithm has been evaluated on an embedded system (Jetson AGX Xavier), which demonstrates the usefulness of this method on power-constrained devices. The proposed algorithm is fast, memory efficient, and computationally less expensive to achieve high detection performance. It is expected to contribute significantly to the wider use of UAV applications including search and rescue missions to locate missing people, and surveillance particularly for applications running on resource-constrained platforms, like smartphones or tablets. This proposed system is now being used in aerial drone system of Police of Scotland to help them locate and find missing and vulnerable people. The results of the project were broadcasted by BBC Scotland.

Taking the bait: The influence of attractants and microhabitat on detections of fauna by remote‐sensing cameras

SummaryAutonomously triggered cameras are a common wildlife survey technique. The use of attractants and surrounding microhabitats is likely to influence detection probabilities and survey outcomes; however, few studies consider these factors. We compared three attractants (peanut butter‐based, tuna‐based and a control) in a Latin square design through a coastal shrubland with high microhabitat variability at Cape Otway, Victoria, Australia (38º50ʹS, 143º30ʹE). Deployments involved 36 cameras for four days in each of five years. The percentage cover of each vegetation structural type (low [no or sparse cover], moderate [grass] or high [shrubs]) within 20 m of each camera was calculated and reduced to a single variable using PCA. Dynamic occupancy modelling, with lure type and vegetation structure as covariates of detection probability, found that peanut butter attracted the greatest diversity of species (24 of 35 species, 69%) and yielded the greatest number of detections (50% of 319) when compared with tuna oil (66% and 24%, respectively) and the control (43% and 26%, respectively). Peanut butter attracted more Macropodidae (wallabies) and Muridae (rats and mice); however, vegetation structural variables were the greatest influence on Corvidae/Artamidae (raven/currawong) detections with higher detectability in more open areas. Vegetation structure also influenced Muridae detections. This study reinforces the critical choice of appropriate attractants and camera placement when investigating vertebrate groups and highlights the role of microhabitat in the detection of small mammals and birds. We suggest future large‐scale camera surveys consider different bait types and microhabitats in their designs, to control for any biases and enable future advice on ‘optimal’ methods.

Novel greedy grid-voting algorithm for optimisation placement of multi-camera

The optimal placement of surveillance cameras and determining the configuration of exposure routes to ensure optimal coverage is essentially a combinatorial optimisation problem. In this paper, we proposed two approaches to identify and locate the location of cameras in an area of interest with planning already done at the site. The major contribution of the paper is to propose the modification on the greedy grid voting algorithm, which can control the overlap between each camera's coverage based on the application requirements. The greedy grid-voting algorithm does not follow any execution order for optimisation purposes. We also modified the traditional global greedy algorithm and it produced better coverage of the area of interest compared to the greedy approaches used to calculate the maximum coverage. The results obtained tell us that the approach ensures total coverage with a minimum of sensors than other techniques exist in the literature.

Pure Laparoscopic Donor Right Hepatectomy (PLDRH) - Maneuvers to Achieve Adequate Exposure with Technical Tips and Tricks

Pure Laparoscopic Donor Right Hepatectomy is a technically challenging surgery requiring long hours of focussed team effort. Access and mobilisation:•Placement of camera ports as high as possible in the line of hilum, right subcostal region•Goldfinger/Snake Liver Retractor 5mm via epigastric port for adequate right lobe liver retraction•Use of endoloop for traction of Gall bladder and divided end of falciform ligament. Hilar dissection:•Gold finger passed anterior to the cava and brought out in the space created between the right hepatic vein and middle hepatic vein in the suprahepatic space.•Umbilical tape passed in the space created by the gold finger which helps achieve hanging, useful for parenchymal transection .•ICG guided Liver parenchymal surface marking for accurate parenchymal transection Parenchymal transection:•Use of rubber band traction for sustained and firm liver traction.•Use of CUSA excel (CUSA with monopolar cautery in the same device) for adequate hemostasis while parenchymal transection•Gold finger is used to encircle the entire right portal pedicle, and umbilical tape is brought in this space which helps in completing the remaining parenchymal transection by hanging manoeuvre. Organ procurement via pfannensteil incision:•Right hepatic duct is encircled with hilar sheath using ICG guidance and divided with scissors sharply after placing a clip to the right hepatic duct orifice.•Right hepatic artery clipped, right portal vein stapled and divided while the right hepatic vein and IVC ligament is divided using a stapler.

Animal detections increase by using a wide-angle camera trap model but not by periodically repositioning camera traps within study sites

When using camera traps for wildlife studies, determining suitable camera models and deployment methods is essential for achieving study objectives. We aimed to determine if camera trap performance can be increased by (1) using cameras with wider detection angles, and (2) by periodically repositioning cameras within sites. We compared three camera trap groups: stationary Reconyx PC900/HC600 (40° detection angle), and paired, periodically-repositioned Reconyx PC900/HC600 and Swift 3C wide-angle camera traps (110° detection angle). Cameras operated simultaneously at 17 sites over 9 weeks within the Upper Warren region, Western Australia. Swift cameras had significantly higher detection rates, leading to better performance, especially for species <1 kg and >10 kg bodyweight. Reconyx cameras missed 54% of known events, with most being animals that moved within the cameras’ detection zones. Stationary and periodically-repositioned Reconyx camera traps performed similarly, although there were notable differences for some species. The better performance of Swift 3C wide-angle camera traps makes them more useful for community-level and species-level studies. The increased sensitivity of the Swift’s passive infrared sensor along with the wider detection zone played an important role in its success. When choosing camera trap models, detection angle and sensor sensitivity should be considered to produce reliable study results. Periodically repositioning cameras within sites is a technique that warrants further investigation as it may reduce camera placement bias, animal avoidance of camera traps, and increase spatial/habitat information when a limited number of cameras are deployed.

CCTV placement in Gaborone City, Botswana

Public surveillance technology through the installation of Closed-Circuit Television cameras (CCTV) has been widely acknowledged as a tool for monitoring population movements and preventing crime. Based on this technological value, the installation of CCTV cameras has become a growing trend in many cities globally. The year 2018 will be remembered as the year when CCTV cameras were first installed in the city of Gaborone, Botswana, and its surrounding villages for purposes of detecting criminal activities and preventing crime. The value of CCTV cameras in preventing crime and as an investigative tool has been an area of interest among researchers. Among scholarly studies on this field, the focus has been on the effectiveness of CCTV cameras in preventing crime and their value as an investigative tool. Since CCTV cameras have just been installed in the city of Gaborone, it may be too early to evaluate the extent to which they effectively prevent crime in this city. The purpose of this study is to document the use of public surveillance cameras in Gaborone and its surrounding areas and assess their geographic placement in light of the principles of Situational Crime Prevention (SCP) theory. Using data collected through site observation and key informant interviews, we argue for a rigorous review and assessment of the current installation and placement of CCTV cameras in Gaborone city and further scholarly study to measure and evaluate the effectiveness of CCTV camera use for crime prevention in this city.

PSIX-37 Considerations for using game cameras for on-farm behavior of beef cattle

Abstract On-farm studies provide an opportunity to build relationships with clientele while answering applied science questions. Further, game trail cameras are a low-cost technology that may be an alternative to visually observing livestock for behavior data. An on-farm study involving eight county Extension agents and producers investigated game cameras to capture cattle visitation to mineral feeders. Three different models of Browning game trail cameras (Prometheus Group LLC, Birmingham, AL) each with timelapse feature were utilized. A three-page document provided background and methodology for the study. Details for placement of the trail camera were provided to provide consistency across farms. Cameras were pre-programmed to settings for image capture. A video for camera setup was also provided in the event that cameras reset to default settings due to battery failure. Timelapse feature was utilized providing image capture from sunrise to sunset at 2 min intervals. Several observations were noted to improve data capture in the future. Differences existed in battery utilization between camera models with one model only capturing images for approximately 2 of the 7 d periods while others had more than 80% battery life remaining after 7 d. Cameras were to be oriented facing south to avoid sun glare. Sun glare occurred at sunrise on most cameras and was particularly bad when dew was present on the camera lens. Mineral feeders need to be secured in place as livestock may move feeders out of the frame of the images captured. Still photos are insufficient in assessing behavior and short video clips are expected to improve interpretation especially when multiple animals visit the feeder at the same time. Utilization of human observation was not useful unless observation was performed at sunrise. Game trail cameras have the potential to be a cost-effective tool in capturing on-farm data of cattle behavior.

Visual Analytics for Operation-Level Construction Monitoring and Documentation: State-of-the-Art Technologies, Research Challenges, and Future Directions

Operation-level vision-based monitoring and documentation has drawn significant attention from construction practitioners and researchers. To automate the operation-level monitoring of construction and built environments, there have been much effort to develop computer vision technologies. Despite their encouraging findings, it remains a major challenge to exploit technologies in real construction projects, implying that there are knowledge gaps in practice and theory. To fill such knowledge gaps, this study thoroughly reviews 119 papers on operation-level vision-based construction monitoring, published in mainstream construction informatics journals. Existing research papers can be categorized into three sequential technologies: (1) camera placement for operation-level construction monitoring, (2) single-camera-based construction monitoring and documentation, and (3) multi-camera-based onsite information integration and construction monitoring. For each technology, state-of-the-art algorithms, open challenges, and future directions are discussed.

Deep invariant texture features for water image classification

Detecting potential issues in naturally captured images of water is a challenging task due to visual similarities between clean and polluted water, as well as causes posed by image acquisition with different camera angles and placements. This paper presents novel deep invariant texture features along with a deep network for detecting clean and polluted water images. The proposed method first divides an input image into H, S and V components to extract finer details. For each of the color spaces, the proposed approach generates two directional coherence images based on Eigen value analysis and gradient distribution, which results in enhanced images. Then the proposed method extracts scale invariant gradient orientations based on Gaussian first order derivative filters on different standard deviations to study texture of each smoothed image. To strengthen the above features, we explore the combination of Gabor-wavelet-binary pattern for extracting texture of the input water image. The proposed method integrates merits of aforementioned features and the features extracted by VGG16 deep learning model to obtain a single feature vector. Furthermore, the extracted feature is fed to a gradient boosting decision tree for water image detection. A variety of experimental results on a large dataset containing different types of clean and stagnant water images show that the proposed method outperforms the existing methods in terms of classification rate and accuracy.

How reliable are motion‐triggered camera traps for detecting small mammals and birds in ecological studies?

AbstractEcological studies often require observations of animals and their behaviour. Motion‐activated cameras (camera traps) based on passive infrared detection (PIR) are a popular solution for recording animal activity in situations when it is impractical for humans to make sufficient observations. However, the reliability of these cameras for recording smaller vertebrates remains uncertain. We assessed the reliability of two widely used PIR camera traps (Bushnell 119740 and Moultrie 13068) for detecting small vertebrates. Specifically, we tested the effects of (1) camera trap model, (2) camera–subject distance and (3) animal class and size on the probability of detection. Brown rats moving across the ground were detected by the Bushnell camera with >80% probability at camera–subject distances ranging from 60 cm to 2 m, while the Moultrie camera was less efficient at distances >1 m. The probability of the Bushnell camera detecting birds feeding on flowers decreased from c. 80% at distances of 40–60 cm to <10% at 2 m and beyond. Larger birds (>20 g) were more likely to be detected than smaller birds (<15 g). Close‐focusing lenses on the Bushnell camera readily allow identification of individual bird species. These results help to establish guidelines for camera trap selection and placement in ecological studies of small terrestrial mammals and birds.

James Burrows (Director)

James Burrows (1940–) knows a thing or two about camera placement. The famed sitcom director of Taxi, Cheers, Friends, and Will & Grace is one of the most prolific television directors of all time, with over one thousand episodes of TV on his resume. But Burrows didn’t always aspire to a career in television. The son of playwright/director Abe Burrows, he grew up with an inside look at the entertainment industry and wanted nothing to do with it. He changed his mind in the mid-1960s. Gary Rutkowski conducted the interview on December 17, 2003, in Los Angeles, California.

Estimating Abundance of an Unmarked, Low‐Density Species using Cameras

ABSTRACTEstimating abundance of wildlife populations can be challenging and costly, especially for species that are difficult to detect and that live at low densities, such as cougars (Puma concolor). Remote, motion‐sensitive cameras are a relatively efficient monitoring tool, but most abundance estimation techniques using remote cameras rely on some or all of the population being uniquely identifiable. Recently developed methods estimate abundance from encounter rates with remote cameras and do not require identifiable individuals. We used 2 methods, the time‐to‐event and space‐to‐event models, to estimate the density of 2 cougar populations in Idaho, USA, over 3 winters from 2016–2019. We concurrently estimated cougar density using the random encounter model (REM), an existing camera‐based method for unmarked populations, and genetic spatial capture recapture (SCR), an established method for monitoring cougar populations. In surveys for which we successfully estimated density using the SCR model, the time‐to‐event estimates were more precise and showed comparable variation between survey years. The space‐to‐event estimates were less precise than the SCR estimates and were more variable between survey years. Compared to REM, time‐to‐event was more precise and consistent, and space‐to‐event was less precise and consistent. Low sample sizes made the space‐to‐event and SCR models inconsistent from survey to survey, and non‐random camera placement may have biased both of the camera‐based estimators high. We show that camera‐based estimators can perform comparably to existing methods for estimating abundance in unmarked species that live at low densities. With the time‐ and space‐to‐event models, managers could use remote cameras to monitor populations of multiple species at broader spatial and temporal scales than existing methods allow. © 2020 The Wildlife Society.

Maximizing Camera Coverage in Multicamera Surveillance Networks

Efficient placement of cameras to perform surveillance tasks has a significant impact on the overall performance as well as cost of video surveillance systems. Identifying an optimal configuration for cameras in a surveillance system to maximize coverage is a combinatorial optimization problem. This paper proposes two algorithms for identifying optimal camera configuration for a multi-camera network with predefined camera locations. We propose an Alternate Global Greedy (AGG) algorithm, a novel variant of the Global Greedy algorithm by making significant modifications to the traditional execution strategy to produce better coverage results. Additionally, we propose an innovative Greedy Grid Voting (GGV) algorithm, which gives preference to cover unique as well as critical regions and effectively addresses a broad range of coverage scenarios mandated by the application. The proposed algorithms are validated on map images across diverse scenarios with predefined camera locations. Experimental results show that our algorithms maximize the coverage results with minimal overlap, compared to the existing greedy techniques.

SCENARIO MODEL TO REDUCE TRAFFIC CONGESTION USING INTELLIGENT TRANSPORTATION SYSTEMS

This study aims to develop a scenario model to reduce traffic congestion using intelligent transportation systems (ITS). ITS is an application of advanced technology in the fields of electronics, computers, and telecommunications to make transportation infrastructure and facilities more informative, smooth, safe, and comfortable. It encompasses the latest wireless, electronic, simulation, and automatic technology. As a method to develop the scenario model, system dynamics (SD) is utilized considering that it can accommodate a system with complex non-linearity, faster and easier sensitivity analyses through the tests on the structure of the models. SD has been used at the macroscopic and microscopic levels of the traffic flow to explore the interaction of transportation and urban planning as well as to evaluate the effect of different transport policies. ITS can reduce traffic congestion through the placement of several surveillance cameras in some corners of the city and is equipped with sensors to detect the number of vehicles. This detection results can be used as an input in setting the time for traffic signal control to reduce the volume of vehicles by prioritizing solid lines to get the green light so that the flow of heavy traffic can run first. With this scenario, traffic congestion is projected to decrease to be in the range of 0.71 – 0.79 (below the maximum saturation level of 0.85) due to the decrease in vehicle volume as the impact of the implementation of ITS.

Vehicle Detection with Self-Training for Adaptative Video Processing Embedded Platform

Traffic monitoring from closed-circuit television (CCTV) cameras on embedded systems is the subject of the performed experiments. Solving this problem encounters difficulties related to the hardware limitations, and possible camera placement in various positions which affects the system performance. To satisfy the hardware requirements, vehicle detection is performed using a lightweight Convolutional Neural Network (CNN), named SqueezeDet, while, for tracking, the Simple Online and Realtime Tracking (SORT) algorithm is applied, allowing for real-time processing on an NVIDIA Jetson Tx2. To allow for adaptation of the system to the deployment environment, a procedure was implemented leading to generating labels in an unsupervised manner with the help of background modelling and the tracking algorithm. The acquired labels are further used for fine-tuning the model, resulting in a meaningful increase in the traffic estimation accuracy, and moreover, adding only minimal human effort to the process allows for further accuracy improvement. The proposed methods, and the results of experiments organised under real-world test conditions are presented in the paper.

32 Web-based Camera (NICView) as a sense of proximity tool: a Quality-Improvement initiative for parents of neonates admitted in the NICU

Abstract Background Family-centered care (FCC) is an important approach and a vital component in healthcare. However, implementation of FCC model in a clinical setting presents a challenge. Effective parent-health care provider relationships are very essential in NICU especially when treatment is complex during the management of critically-ill neonates. Web-based camera (WBC) systems have been used in a few NICUs in the USA and the UK, allowing virtual visitation of the family members. However, to the best of our knowledge, there is no center in Canada that has utilized this WBC (NICView ®) service. NICView® is a WBC mounted to the baby’s bed, which is streaming ‘live’ video around-the-clock to parents and families. We report the perception of parents and their satisfaction after using NICView® facility in our NICU. Objectives Our goal was to see if the WBC system reduced the anxiety and stress level in parents by exploring NICView ® as a tool to enhance FCC model. This study was designed to explore the perception of parents and their satisfaction levels during their newborns stay within our NICU. Design/Methods A prospective, questionnaire-based study conducted over 6-month period. Determinants that may facilitate or disrupt the use of a commercially-available camera device were identified. After signing an informed consent form, the parents utilized the camera service continuously throughout the newborn stay within the NICU, except during the times of procedures, feedings and routine baby care. Demographic information of parents was collected in the questionnaire, along with along satisfaction questions. Parents filled out the questionnaires after 1-week of utilization of the camera or sooner if the baby was getting discharged. Results 94 respondents completed the questionnaire. Parental perception was consistent with earlier studies. Almost all parents used the service on a regular basis, twenty NICView cameras had more than 7000 logins by respondents during study period. Family members residing out of Saskatoon or out-of-SK and in some instances out-of-Canada used it the most, especially working fathers or distant grandparents. Seventy-five percent of the respondents were mothers. Almost 89% of parents reported having the WBC system helped reduce their anxiety and stress levels. All parents (100%) recommended this WBC system to other parents and strongly agreed that NICView was easy to use. Nearly 98% of parents agreed that they felt very connected with their baby using the WBC system. Frequently reported comments were on the positioning of camera like incorrect camera placement that did not offer a good view of the baby, the need to call nurses sometimes to turn on or adjust the camera after care, using the chat option to adjust the camera position when moved away. After we saw these frequent comments, we had to reiterate to bed-side nurses to re-position the camera if moved away and reminded parents to make use of the NICView® 24-hr customer support service available offered by the company. Conclusion