Static Camera Research Articles

Method of Moving Region Detection for Static Camera

The moving object detection from a stationary video sequence is a primary task in various computer vision applications. In this proposed system; three processing levels are suppose to perform: detects moving objects region from the background image; reduce noise from the pixels of detected region and extract meaningful objects and their features (area of object, center point of area etc.). In this paper; background subtraction techniques is used for segments moving objects from the background image, which is capable for pixel level processing. Morphology operation (Erosion and dilation) are used to remove pixel to pixel noise. In last level, CCL algorithm is used for sorts out foregrounds pixels are grouped into meaningful connected regions and their features.

Background Modeling and Referencing for Moving Cameras-Captured Surveillance Video Coding in HEVC

Surveillance video coding is crucial for improving compression efficiency in intelligent video surveillance systems and applications. Plenty of work has been done, which can be roughly divided into two categories: the former mainly focuses on low-complexity background modeling to obtain the clear background, while the latter focuses on an appropriate coding strategy to generate the high-quality background reference picture for effective background prediction. However, almost all existing works focus only on stationary camera scenes, while moving cameras-captured surveillance video coding is left untouched and is still an open problem. In this paper, a background modeling and referencing scheme for moving cameras-captured surveillance video coding in high-efficiency video coding (HEVC) is proposed. First, this paper proposes a low-complexity motion background modeling algorithm for surveillance video coding using the running average based on a global-motion-compensation method. To obtain the global motion vector, we propose a global motion detection method based on character blocks by establishing a low-rank singular value decomposition model for clustering and estimating motion vectors of background character blocks in the cameras movement circumstance. Second, we propose a background referencing coding strategy, in which the motion background coding tree units (MBCTUs) would be selected by anchoring the input video frame on the modeling background frame and coded with the optimized quantization parameter. Then, the reconstructed MBCTU will be used to update the previous coding tree unit in the global compensation location of the background reference picture. Extensive experimental results show that the proposed scheme can achieve significant bit savings of up to 26.6% and, on average, 6.7% with similar subjective quality and negligible encoding complexity, compared to HM12.0. Besides, the proposed scheme consistently outperforms two state-of-the-art surveillance video coding schemes with remarkable bitrate savings.

Vehicle Detection by Fusing Part Model Learning and Semantic Scene Information for Complex Urban Surveillance.

Visual-based vehicle detection has been studied extensively, however there are great challenges in certain settings. To solve this problem, this paper proposes a probabilistic framework combining a scene model with a pattern recognition method for vehicle detection by a stationary camera. A semisupervised viewpoint inference method is proposed in which five viewpoints are defined. For a specific monitoring scene, the vehicle motion pattern corresponding to road structures is obtained by using trajectory clustering through an offline procedure. Then, the possible vehicle location and the probability distribution around the viewpoint in a fixed location are calculated. For each viewpoint, the vehicle model described by a deformable part model (DPM) and a conditional random field (CRF) is learned. Scores of root and parts and their spatial configuration generated by the DPM are used to learn the CRF model. The occlusion states of vehicles are defined based on the visibility of their parts and considered as latent variables in the CRF. In the online procedure, the output of the CRF, which is considered as an adjusted vehicle detection result compared with the DPM, is combined with the probability of the apparent viewpoint in a location to give the final vehicle detection result. Quantitative experiments under a variety of traffic conditions have been contrasted to test our method. The experimental results illustrate that our method performs well and is able to deal with various vehicle viewpoints and shapes effectively. In particular, our approach performs well in complex traffic conditions with vehicle occlusion.

Detection of aeolian transport in coastal images

Abstract Optical remote sensing provides a low cost method of documenting surface conditions over extended periods of time. Its utility in resolving large scale aeolian transport events is examined in this paper. To this end, over 1800 images taken over a three month period by a stationary camera in a coastal setting are classified as containing/ not containing aeolian transport using both a manual and automated approach. In both cases transport is inferred by the identification of aeolian streamers within a given image. With regards to the manual approach, images were visually examined by three users. There was good agreement between the different users with 77.8% of transport identifications being unanimous. Comparison of the user results with measured transport highlights that observations of transport in images coincide with increased particle counts. The particle counts indicate that transport is dominated by a series of discernible transport events. Transport is observed in images throughout the duration of these events. Manually processing the images is a laborious process. A method is presented which allows for the automated detection of images containing transport. The method utilises 2D-DFTs to exploit the linear appearance of streamers in images. Relative to the manual classifications which serve as ground truth, the method performs well in identifying images containing transport with values of precision, recall and F1 score ⩾ 0.85 .

Breaking the Limits in Urban Video Monitoring: Massive Crowd Sourced Surveillance over Vehicles

Contemporary urban environments are in prompt need of the means for intelligent decision- making, where a crucial role belongs to smart video surveillance systems. While existing deployments of stationary monitoring cameras already deliver notable societal benefits, the proposed concept of massive video surveillance over connected vehicles that we contribute in this article may further augment these important capabilities. We therefore introduce the envisioned system concept, discuss its implementation, outline the high-level architecture, and identify major data flows, while also offering insights into the corresponding design and deployment aspects. Our conducted case study confirms the potential of the described crowd sourced vehicular system to effectively complement and eventually surpass even the best of today's static video surveillance setups. We expect that our proposal will become of value and integrate seamlessly into the future Internetof- Things landscape, thus enabling a plethora of advanced urban applications.

Motion-Based Object Segmentation Based on Dense RGB-D Scene Flow

Given two consecutive RGB-D images, we propose a model that estimates a dense 3D motion field, also known as scene flow. We take advantage of the fact that in robot manipulation scenarios, scenes often consist of a set of rigidly moving objects. Our model jointly estimates (i) the segmentation of the scene into an unknown but finite number of objects, (ii) the motion trajectories of these objects and (iii) the object scene flow. We employ an hourglass, deep neural network architecture. In the encoding stage, the RGB and depth images undergo spatial compression and correlation. In the decoding stage, the model outputs three images containing a per-pixel estimate of the corresponding object center as well as object translation and rotation. This forms the basis for inferring the object segmentation and final object scene flow. To evaluate our model, we generated a new and challenging, large-scale, synthetic dataset that is specifically targeted at robotic manipulation: It contains a large number of scenes with a very diverse set of simultaneously moving 3D objects and is recorded with a simulated, static RGB-D camera. In quantitative experiments, we show that we outperform state-of-the-art scene flow and motion-segmentation methods on this data set. In qualitative experiments, we show how our learned model transfers to challenging real-world scenes, visually generating better results than existing methods.

LSPIV Measurements of Two‐Dimensional Flow Structure in Streams Using Small Unmanned Aerial Systems: 1. Accuracy Assessment Based on Comparison With Stationary Camera Platforms and In‐Stream Velocity Measurements

AbstractMeasuring two‐dimensional (2‐D) patterns of flow in rivers at high resolution over large areas is challenging using traditional velocity‐measurement methods, which provide data at specific locations or cross sections. Large‐scale particle image velocimetry (LSPIV) based on imagery obtained from fixed camera platforms can measure flow velocity on the surface of rivers and is generally accurate compared to near‐surface velocity measurements obtained by traditional methods. The proliferation of inexpensive small unmanned aerial systems (sUAS) equipped with high‐resolution cameras and onboard GPS has the potential to facilitate measurements of flow patterns in rivers using LSPIV, but few studies have assessed the accuracy of sUAS‐derived LSPIV compared to fixed‐platform LSPIV and in‐stream velocity measurements. This study assesses the accuracy of sUAS‐based LSPIV for measuring 2‐D mean surface velocities as well as quasi‐instantaneous 2‐D velocities obtained from successive image frames. For persistent 2‐D flow, mean velocities derived from sUAS‐based LSPIV match those obtained by stationary camera platforms, and velocities measured by both LSPIV methods agree with near‐surface velocities measured by an acoustic Doppler velocimeter. Quasi‐instantaneous velocities are degraded by camera movement and low pixel resolution, but capturing the evolution of 2‐D flow structures is possible in certain circumstances. The results confirm that sUAS‐derived LSPIV provides accurate, high‐resolution measurements of mean surface velocities over large spatial areas of persistent 2‐D flow and can characterize evolving 2‐D flow structures under favorable conditions. sUAS‐based LSPIV is a valuable new method for mapping of 2‐D patterns of surface flow in rivers—an issue explored in a companion paper.

Automatic Semantic Parsing of the Ground Plane in Scenarios Recorded With Multiple Moving Cameras

Nowadays, video surveillance scenarios usually rely on manually annotated focus areas to constrain automatic video analysis tasks. Although manual annotation simplifies several stages of the analysis, its use hinders the scalability of the developed solutions and might induce operational problems in scenarios recorded with multiple moving cameras (MMCs). To tackle these problems, an automatic method for the cooperative extraction of areas of interest (AoIs) is proposed. Each captured frame is segmented into regions with semantic roles using a state-of-the-art method. Semantic evidences from different junctures, cameras, and points-of-view are, then, spatio-temporally aligned on a common ground plane. Experimental results on widely used datasets recorded with multiple but static cameras suggest that this process provides broader and more accurate AoIs than those manually defined in the datasets. Moreover, the proposed method naturally determines the projection of obstacles and functional objects in the scene, paving the road towards systems focused on the automatic analysis of human behavior. To our knowledge, this is the first study dealing with this problem, as evidenced by the lack of publicly available MMC benchmarks. To also cope with this issue, we provide a new MMC dataset with associated semantic scene annotations.

A Novel Algorithm Based on the Pixel-Entropy for Automatic Detection of Number of Lanes, Lane Centers, and Lane Division Lines Formation.

Lane detection for traffic surveillance in intelligent transportation systems is a challenge for vision-based systems. In this paper, a novel pixel-entropy based algorithm for the automatic detection of the number of lanes and their centers, as well as the formation of their division lines is proposed. Using as input a video from a static camera, each pixel behavior in the gray color space is modeled by a time series; then, for a time period , its histogram followed by its entropy are calculated. Three different types of theoretical pixel-entropy behaviors can be distinguished: (1) the pixel-entropy at the lane center shows a high value; (2) the pixel-entropy at the lane division line shows a low value; and (3) a pixel not belonging to the road has an entropy value close to zero. From the road video, several small rectangle areas are captured, each with only a few full rows of pixels. For each pixel of these areas, the entropy is calculated, then for each area or row an entropy curve is produced, which, when smoothed, has as many local maxima as lanes and one more local minima than lane division lines. For the purpose of testing, several real traffic scenarios under different weather conditions with other moving objects were used. However, these background objects, which are out of road, were filtered out. Our algorithm, compared to others based on trajectories of vehicles, shows the following advantages: (1) the lowest computational time for lane detection (only 32 s with a traffic flow of one vehicle/s per-lane); and (2) better results under high traffic flow with congestion and vehicle occlusion. Instead of detecting road markings, it forms lane-dividing lines. Here, the entropies of Shannon and Tsallis were used, but the entropy of Tsallis for a selected q of a finite set achieved the best results.

Application of non-destructive methods for assessing rock pool fish assemblages on Lord Howe Island, Australia

Rock pool fish assemblages can be highly diverse, though are poorly studied in many locations. Where rock pool fishes have been studied, sampling has often been undertaken using destructive techniques, causing fish mortality and damage to pool ecosystems. There is, therefore, a need for greater understanding of non-destructive methods for evaluating rock pool fish assemblages. To improve knowledge in this area, we tested three non-destructive techniques: mini baited remote underwater videos (mini-BRUVs) which utilised a stationary video camera to record fish, visual censuses (VCs) where a roving observer recorded fish using a slate and observer operated videos (OOVs) where fish were recorded by an observer using a video camera.These methods were tested for their effectiveness in assessing fish assemblages, using data from rock pools on Lord Howe Island (LHI), in New South Wales (NSW), Australia. VCs and OOVs required significantly less total survey time than mini-BRUVs, whereas mini-BRUVs provided advantages in the detection of fishes in small pools (<25 m3). Examination of mini-BRUVs’ data, for 30 min samples, identified that the vast majority of fish species were detected within 15 min and indicated that a shorter 15 min sampling period would generally be suitable for collecting fish data from rock pools. The base-line data collected on fishes in LHI rock pools demonstrated the uniqueness of LHI assemblages, which were significantly different from those on the NSW coast. This study demonstrates that non-destructive sampling methods can be used in rock pool environments, making these methods suitable for sampling in inter-tidal habitats, especially within marine protected areas.

A method for computing volumetric fish density using stereo cameras

Cameras are an important tool for sampling marine environments, providing estimates of fish size and abundance for ecological studies and resource management. In addition to length estimates, calibrated stereo-cameras can be used to provide precise locations (i.e. range and angle) of objects within the view frame of both cameras. We present a general method to exploit this information to estimate the density of fishes from stereo-camera measurements. This includes deriving the jointly sampled volume from stereo-camera images, accounting for the presence of the sea floor within the camera view, and accounting for detectability of fishes as a function of range. The method was demonstrated on a set of paired still images collected by stationary camera platforms deployed on the seafloor. Estimates of fish range, camera sampling volume, and seafloor position were used to model volumetric density as a function of range from the camera. A comparison of volumetric density estimates and traditional count metrics found that these metrics differ substantially in inferred relative species composition. The potential value of volumetric density for camera-based marine surveys, the potential for deriving absolute abundance, and the possibility of range-induced bias in count data are discussed.

Video Synchronization Based on Projective-Invariant Descriptor

In this paper, we present a novel trajectory-based method to synchronize two videos shooting the same dynamic scene, which are recorded by stationary un-calibrated cameras from different viewpoints. The core algorithm is carried out in two steps: projective-invariant descriptor construction and trajectory points matching. In the first step, a new five-coplanar-points structure is proposed to compute the cross ratio during the construction of the projective-invariant descriptor. The five points include one trajectory point and four fixed points induced from the background scene, which are co-planar in the 3D coordinate. In the second step, the matched trajectory points are initially estimated by the primitive nearest neighbor method, and are further refined by using epipolar geometric constraints and post processing. Experimental results demonstrate that the proposed method significantly outperforms the existing state-of-the-arts. More importantly, the proposed method is more generic in the sense that it works well for those videos captured under different conditions, including different frame rates, wide baseline, multiple moving objects, planar or non-planar motion trajectories.

The Replate

Replates are a new way to experience replays of sporting events. We use computer vision tools, such as camera tracking and background subtraction, to align all frames in the original video with regard to each other. Using this alignment, we create a single plate as it would be seen from a wide-angle static camera. Key players and events are then reinserted into this arena using an interactive tool, recovering the big picture that is typically lost due to camera panning and zoom. Replates are played in continuous loop. We allow replate authors to include multiple instances of each player, offset by regular intervals, so viewers can quickly and repeatedly inspect their favorite moments. Authors can also choose from a variety of effects that help them construct a narrative of their favourite plays. The interactive system gives immediate feedback allowing authors to adjust the rendered replates interactively. Replates take a fraction of the size of the original input video, and produce an effect reminiscent of video game reenactments of famous sporting events. The footage, however, is real.

Foreground segmentation with PTZ camera: a survey

The alertness of terrorism in the present is greater than that in the past b with reference to the incident of September 11. Still now, there has been a fight against terrorism and that has triggered a novel effort to locate the enhanced approaches with a higher-end camera. A Pan Tilt Zoom (PTZ) camera, which is a type of such high-end camera with multi-functionalities, can be used for identifying such potential threats. Consequently, the background modeling has an increasing significance in the computer vision to segment the foreground objects for further analysis in video surveillance applications. A PTZ camera offers a lot of benefits over normal fixed cameras. It provides an easy installation with 360° plane and greater flexibility. Although numerous surveys on static camera methods have already been proposed to model background, these methods do not adopt maximized large-scale scene coverage as well as frame quality to recognize specific targets compared to the PTZ camera. This motivates the survey to address the issues and techniques related to the PTZ background modeling, since there is no survey on this emerging area. The sole objective of this paper is to present a brief survey on the PTZ camera-based foreground segmentation method, which is very indispensable for high level analysis. It also provides an overview of various techniques from the literature that addresses the challenges, solutions, key aspects of the PTZ camera-based foreground segmentation methods, categorization of different approaches as well as the available datasets used for experimentation, and important future scope along with left over challenges for the computer vision researchers with applications.

A block-wise frame difference method for real-time video motion detection

This article proposes a motion detection method for real-time video analysis. It is the fundamental principle that the parts of the moving objects and the local changes of the images captured by static cameras are strongly correlated. Peak signal-to-noise ratio calculated in a block can characterize the significance of the changes in this area. Moving objects can therefore be detected by thresholding the peak signal-to-noise ratio of the blocks between two adjacent frames. The block-wise scheme used in this frame difference method can explore the local correlation of the movement in both space and time domains. This approach is robust to analyze the video images with noise and high variance caused by environmental changes, such as illuminations changes. Compared with other methods, the proposed method can achieve relatively high detection accuracy with less computation time, where real-time motion detection is available. Experimental results show that the proposed method cost averagely 50% of the running time of ViBe with 3.5% increase of the F-score on detection accuracy.

Identification of a Moving Object's Velocity and Range With a Static-Moving Camera System

In this technical note, a nonlinear observer strategy is proposed to asymptotically identify the velocity and range of the feature points on a moving object using a static-moving camera system. Specifically, the camera system consists of a static camera and a moving camera. The observers are developed in a sequential way. First, nonlinear observers are designed to identify the scaled velocity of the feature points and the angular velocity of the moving camera. Second, based on the estimated velocities, the range of the feature points is identified by a reduced order estimator. Homography-based techniques are used in the auxiliary state construction and the observer design. With the estimated scaled velocity and range, the motion and structure of the object can be obtained with respect to both cameras. The proposed approach does not require the motion constraint or a priori geometric knowledge of the moving object. Furthermore, Lyapunov-based analysis is used to prove that the estimators asymptotically identify the velocity and range of the feature points. The performance of the proposed strategy is demonstrated by simulation results.

Comparison of user satisfaction and image quality of fixed and mobile camera systems for 3-dimensional image capture of edentulous patients: A pilot clinical study

Statement of problemAn evaluation of user satisfaction and image quality of a novel handheld purpose-built mobile camera system for 3-dimensional (3D) facial acquisition is lacking. PurposeThe purpose of this pilot clinical study was to assess and compare the effectiveness between a handheld mobile camera system designed for facial acquisition and a fixed static camera arrangement by comparing the time effectiveness and the operator and participant preference for the 2 techniques of image capture. Material and methodsCompletely edentulous participants (n=12: women=7, men=5; mean age: 74.6 years) were included in this pilot study. Images were captured with and without the prostheses in situ while maintaining “serious” and “full-smile” facial expressions. Images were captured using a mobile and a static system. The working times for the participant installation and image captures were recorded. Operator and participant perceptions of the entire experience were recorded by using visual analog scale questionnaires. Nonparametric tests were used for statistical analyses (α=.05). ResultsThe installation time was significantly shorter for the mobile system (static=24 ±13 seconds; mobile=10 ±10 seconds), but the differences in the image capture times were not statistically significant (static: 29 ±5 seconds; mobile: 40 ±18 seconds). Operator preference was in favor of the mobile system with regard to working time (P=.002), difficulty in using (installation: P=.002; handling: P=.045), and camera weight (P=.002); however, they preferred the static arrangement for image quality (P=.003) and comfort (P=.013). The participants rated the entire photographic experience favorably, and 10 of 12 participants preferred the static camera over the mobile one. ConclusionsDespite the complexity of the installation, the static system was evaluated better for image quality; the mobile system was easier in installation and handling. The operators preferred the mobile system, and the participants preferred the static system.

ICT SYSTEMS FOR MONITORING AND PROTECTION OF WILDLIFE IN THEIR NATURAL ENVIRONMENT

The paper deals with systems for monitoring and protection of wildanimals in their natural environments and the use IoT technologies and solutionsin protected nature reserves. The paper also examines the reasons and possibilitiesfor implementing the above mentioned technical solutions, especially in termsof protecting species from the red list of endangered species. In this sense, thepaper also discusses technological solutions and the possibilities of applying IoTworking framework, the concept of the Internet of animals, and the application ofthese technologies through various business and research models. Finally, the paperprovides examples of solutions from the point of view of the necessary infrastructure(servers, storage, internet, animal necklaces, stationary cameras and drones), aswell as from the point of view of data processing and the legal framework for theapplication of these solutions.

Full-view barrier coverage in mobile camera sensor networks

Barrier coverage is an important problem in wireless camera sensor networks for many security applications such as border monitoring and target detection, where camera sensors are used for detecting targets that penetrate the protected area. The full-view coverage is an effective approach of camera barrier coverage which can capture multiple viewpoints of the target crossing the protected area. Existing works on full-view barrier coverage mainly consider static and rotatable camera sensors, which leads to a waste of camera sensors under the limitation of sensing angles and coverage areas. In this paper, we study full-view barrier coverage in mobile camera sensor networks, where camera sensors can randomly move within the deployed area. Specially, we first propose full-view covered model of mobile camera sensors. With this model, we then divide the deployed area into some connected grids and present grid-based deployment strategy to deploy camera sensors for each grid. Finally, we construct a weighted directed graph to model these connected grids and their relationship. Based on the graph, we employ Dijkstras algorithm to obtain a shortest coverage barrier, which is a connected full-view coverage zone across the entire target area. Extensive simulation and theoretical results are provided to illustrate the efficient of our covered models as well as our proposed algorithm in reducing the number of camera sensors and improving coverage probability.

Target Tracking in the Presence of Intermittent Measurements via Motion Model Learning

When using a camera to estimate the pose of a moving target, measurements may only be available intermittently, due to feature tracking losses from occlusions or the limited field of view of the camera. Results spanning back to the Kalman filter have demonstrated the utility of using a predictor to update state estimates when measurements are not available, but target velocity measurements or a motion model must be known to implement a predictor for image-based pose estimation. In this paper, a novel estimator and predictor are developed to simultaneously learn a motion model, and estimate the pose, of a moving target from a moving camera. A stability analysis is provided to prove convergence of the state estimates and function approximation without requiring the restrictive persistent excitation condition. Two experiments illustrate the performance of the developed estimator and predictor. One experiment involves a stationary camera observing a mobile robot with sporadic feature tracking losses, and a second experiment involves a quadcopter moving between two mobile robots on a road network.