Video Camera Research Articles

Computer vision recognition in the teaching classroom: A Review

Artificial intelligence introduces computer vision recognition into the teaching classroom, and computer vision recognition technology lays a solid foundation for the intelligent teaching classroom. Through the classroom camera video stream to the classroom student information data collection, voice, posture, facial, physiological signal data recognition analysis processing to extract and define the characteristics of student behaviour, automatic classification behaviour and then record and display student behaviour, thus effectively help teachers to grasp the students learning state and emotions, to promote the quality of teaching has far-reaching significance.

AIPT: Adaptive information perception for online multi-object tracking

Information perception is crucial in MOT tasks. Recent approaches use positional, motion, and appearance information to model object states. However, in scenes involving camera motion, tracking tasks suffer from image distortion, trajectory loss, and mismatching issues. In this paper, we propose Adaptive Information Perception for Online Multi-Object Tracking, abbreviated as AIPT. AIPT consists of an Adaptive Motion Perception Module (AMPM) and an Asymmetric Information Suppression Module (AISM). In AMPM, we design an Adaptive Image Distortion Recovery Module (AIDRM) to perceive distortions in unknown scenes, allowing the tracker to autonomously recover distorted images as the scene changes. By designing the Information-Guided Trajectory Restoration Module (IGTRM), the tracker learns object motion states from prior information and constructs accurate reconstruction information during trajectory loss. Furthermore, our AISM module utilizes masking information to suppress potential relationships between asymmetric objects, thereby enhancing the ability of tracker to handle mismatches. Both AMPM and AISM exhibit excellent scalability, seamlessly integrating with most advanced tracking methods. Ultimately, our AIPT achieves leading performance on multiple benchmark platforms, including MOT17, MOT20, and KITTI.

Robust-to-occlusion machine vision model for predicting quality variables with slow-rate measurements

Efficient control and optimization of processes require fast-rate measurements of process variables. However, certain variables can only be measured at a slow rate due to technical or economic limitations. Video cameras are commonly available in process industry. Taking images regularly may provide valuable insights into the process dynamics. In this paper, a vision model is proposed to provide a fast-rate prediction for the slow-rate measured variables. However, since images are susceptible to various deteriorating factors including environmental sources, an autoencoder-based image inpainting method is used to retrieve the true images. Then, a predictive multirate auto-regressive with eXogenous input (ARX) model is developed using image features and slow-rate measurements. Unknown parameters are estimated using the expectation–maximization (EM) algorithm. A Kalman filter and Rauch–Tung–Striebel smoother are used to determine the posterior distribution of states. The proposed vision model demonstrates promising performance in providing fast-rate predictions of the interface level in a laboratory-scale primary separation cell.

Methodology for Assessing Lung Volumes Based on Recording the Biomechanics of Respiratory Movements

The method for studying the biomechanics of breathing has been developed to determine the volume of inhaled and exhaled air. The technique is based on recording the movements of the walls of the chest and abdomen of a person during breathing. A video camera was used to record movements of the chest and abdominal walls, with a spirometer used as the reference method for measuring lung volumes. According to the proposed technique, studies were conducted among 63 people (age – 18–26 years, body mass index ranged from 18.1–32.6 kg/m2, height – 160–192 cm). Statistical processing of the results showed a strong relationship between the measured volu mes and the relative deviations of the walls of the human chest and abdomen. The developed technique for studying the biomechanics of breathing can be used for indirect assessment of lung volumes.

Edge-Computing-Enabled Abnormal Activity Recognition for Visual Surveillance

Due to the ever increasing number of closed circuit television (CCTV) cameras worldwide, it is the need of the hour to automate the screening of video content. Still, the majority of video content is manually screened to detect some anomalous incidence or activity. Automatic abnormal event detection such as theft, burglary, or accidents may be helpful in many situations. However, there are significant difficulties in processing video data acquired by several cameras at a central location, such as bandwidth, latency, large computing resource needs, and so on. To address this issue, an edge-based visual surveillance technique has been implemented, in which video analytics are performed on the edge nodes to detect aberrant incidents in the video stream. Various deep learning models were trained to distinguish 13 different categories of aberrant incidences in video. A customized Bi-LSTM model outperforms existing cutting-edge approaches. This approach is used on edge nodes to process video locally. The user can receive analytics reports and notifications. The experimental findings suggest that the proposed system is appropriate for visual surveillance with increased accuracy and lower cost and processing resources.

Simulate This! Virtual Production and Cinematic Education with cineDESK

Due to the increased availability of computing power, hybrid image production – combining live-action image components with virtual assets – can now be performed in real-time, enabling filmmakers to seamlessly visualize, manipulate and incorporate digital assets into their shooting process. The related novel workflows are commonly referred to as “Virtual Production”. Here, we focus on the use of Virtual Production techniques for cinematic simulation. In 2020, the Zurich University of the Arts created cineDESK, a simulation tool to facilitate the previsualization of film scenes and stimulate the artistic choices of filmmakers. Using cineDESK, directors, cinematographers and production designers can collaboratively explore how space, light, props and acting are translated into cinematic sequences. In educational settings, students benefit from cineDESK’s capacity to explore specific aspects of the filmmaking process in isolation. cineDESK allows them to repeatedly rework their scenes, rearranging and optimizing camera angles and movements, lighting, production design elements and even the appearance of their (virtual) actors. Thus, cineDESK helps filmmakers to develop an idiosyncratic visual language and make informed aesthetic choices without being overwhelmed by the complexities of a real film shoot.

GLOCAL: A self-supervised learning framework for global and local motion estimation

Motions in videos are typically a mixture of local dynamic object motions and global camera motion, which are inconsistent in some cases, and even interfere with each other, causing difficulties in various downstream applications, such as video stabilization that requires the global motion, and action recognition that consumes local motions. Therefore, it is crucial to estimate them separately. Existing methods separate two motions from the mixed motion fields, such as optical flow. However, the quality of mixed motion determines the higher bounds of the performance. In this work, we propose a framework, GLOCAL, to directly estimate global and local motions simultaneously from adjacent frames in a self-supervised manner. Our GLOCAL consists of a Global Motion Estimation (GME) module and a Local Motion Estimation (LME) module. The GME module involves a mixed motion estimation backbone, an implicit bottleneck structure for feature dimension reduction, and an explicit bottleneck for global motion recovery based on the global motion bases with foreground mask under the training guidance of proposed global reconstruction loss. An attention U-Net is adopted for LME which produces local motions while excluding motion of irrelevant regions under the guidance of proposed local reconstruction loss. Our method can achieve better performance than the existing methods on the homography estimation dataset DHE and the action recognition dataset NCAA and UCF-101.

Capture of zooplankton by site-attached fish: striking dynamics under different flow speeds and prey paths

Consumption of pelagic zooplankton plays a vital role in the functioning of benthic communities such as coral reefs and kelp forests. Many fish that consume zooplankton in those habitats are site attached, foraging for drifting prey while maintaining a fixed position close to a shelter such as a branching coral or a perforated rock. Therefore, the flow, in which their planktonic prey drifts, is expected to affect their foraging movements. However, most attributes of those movements are poorly understood- a gap that our study seeks to fulfil. Our experiments were carried out in a laboratory flume with 4 common coral-reef site-attached species. Their movements were recorded in 3D, using two orthogonal video cameras. Different fishes exhibited similar trends despite noticeable differences in their body size, their morphology, the type of shelters they use, and the typical size of the groups in which they reside. In all species, the strike distance decreased with increasing flow speed. Similarly, the distance between the fish and prey at the moment of strike initiation (“Reactive Distance”) decreased with increasing flow speed, as well as the angle between that “Reactive Distance” and flow direction. Surprisingly, striking speeds (relative to Earth) remained nearly unchanged under different flows speeds. However, faster strikes occurred when oriented at wider angles relative to the flow. Taken together, the fish appear to determine the speed and angle of their strikes based on a cognitive ability to assess the prey’s drifting speed and path in order to reach on time the intercepting point. A rough estimate of the time it takes the fish to decide on the strike’s orientation and speed, would suggest a few hundred of milliseconds. Using published data on the fishes’ feeding rates, we found that the fish significantly differed in their feeding efficiencies, defined as the percent of prey they captured from those passing through their actual foraging space. That difference may explain inter-specific differences in the habitats the fish use and their group size.

Spatial and temporal consistency learning for monocular 6D pose estimation

Monocular 6D pose estimation is a challenging task in the field of computer vision and robotics. Many previous works only input the cropped image of single object during training and inference, aiming to remove the noise from non-object regions. However, most of these methods ignore the viewpoint and spatial relationships of objects in the scene, which are crucial for accurate pose estimation of camera. To address this issue, this paper proposes a novel multi-view and multi-object based learning strategy for monocular 6D pose estimation, which involves the consistency of object coordinate for the same object at different viewpoints and the consistency of world coordinate for different objects in the same space. In the proposed method, the spatial and temporal groups are generated to trained the monocular 6D pose estimation network. Due to the camera motion, scene images taken at different times can be regarded as images captured from different viewpoints. Therefore, a temporal consistency loss is designed to constraint the relationship of the same object at different viewpoints, while a spatial consistency loss is designed to constraint the relationship of different objects at the same space. Finally, the proposed method is verified on the public datasets. Experimental results show that the proposed method is accurate, robust, and outperforms similar state-of-the-art approaches.

DANCE FILM FOR THE NEEDS OF THE NATIONAL STUDENT ARTS COMPETITION FESTIVAL (FLS2N) AT SMP NEGERI 3 DEPOK

FLS2N is a forum for students to develop interests and talents in the field of arts and culture and compete to reach the peak of achievement. Through this program, it is hoped that students will be able to develop ideas and creativity in the arts from an early age to create self-confidence in their abilities and explore various potential intelligence. The 2023 FLS2N dance competition will be held by recording dance material in the form of a dance film with various provisions in accordance with the guidelines for recording dance films. In this case, we will review the Leuwi Abyudaya Dance Film created by the PKM PM Team as a form of community service for SMP Negeri 3 Depok. The author uses the creation method according to Alma M Hawkins, where this method consists of the Sensing, Feeling, Imaging, Transforming, Forming process, and use the cinematography composition method by Kurt Lancaster that consist of shot sizes and lenses, camera height and angle, camera motion, focal depth of field, light and dark, line and linear perspective, layers, weight, color, dan texture shot sizes and lenses, camera height and angle, camera motion, focal depth of field, light and dark, line and linear perspective, layers, weight, color, dan texture.

The Effectiveness of Cognitive Behavior Group Counseling Restructuring Technique to Reduce the Social Anxiety of Cyber-Bullying Victim

Bullying is one of the apprehensive problems to the students at school. Bullying does not happens only at school but it also happens at cyberspace and it is called cyber-bullying. This aggression involves the use of technology like mobile phones, video camera, email, and web pages to send the message that humiliate other people. Besides that, platforms like social media which are very favored by many people today, can be a tool to do cyber-bullying. The cyber-bullying victim can get many emotional effects like anxiety, stress, depression, or suicide. The purpose of this study was to determine the effectiveness of cognitive behavioral group counseling using cognitive restructuring techniques in reducing cyber-bullying social anxiety among students. This research uses an experimental approach one group pretest-posttest control design. The subject of the research are twelve students at a high school named SMA Negeri 7 Mataram, Indonesia. The data collection uses scale social anxiety. The data analysis use one way statistic ANNOVA. The results showed that the application of cognitive behavioral group counseling using cognitive restructuring techniques was effective in reducing cyber-bullying social anxiety in students.
 Keywords: cognitive behavior group counseling, cognitive restructuring, cyber-bullying

Robust multiple obstacle tracking method based on depth aware OCSORT for agricultural robots

Multiple Object Tracking (MOT) of dynamic obstacles in field is an important prerequisite for agricultural robots to achieve dynamic obstacle avoidance. The complex and unpredictable road environment in the rural areas will cause severe vibration to the robot, affecting the camera pose and consequently leading to object matching errors. Therefore, we propose an improved method named Depth Aware Observation Centric Simple Online and Realtime Tracking (DA-OCSORT) with two new modules named Inertial Measurement Unit (IMU)-based Camera Motion Compensation (ICMC) and Depth Aware (DA). This method can use IMU information to compensate for camera ego-motion and perform multi-dimensional matching through object depth information, thereby minimizing the influence of camera motion on tracking process. We created an open-source MOT dataset, named MOTorchard. Building upon the pure motion-based method OCSORT, we achieve SOTA on this MOTorchard with HOTA 54.1, MOTA 83.3, IDF1 52.9 and FPS 22.7. Moreover, the proposed two modules prove to have good real-time performance and strong adaptability to other published MOT algorithms.

Immediate effects of circular trunk movements on reach distance and centre of pressure distance during a multidirectional seated reach task in participants with post-stroke hemiparesis: a non-randomised controlled trial

Background/Aims Sitting balance training is an important therapeutic tool used by physical and occupational therapists who work with patients with hemiparesis, many of whom have difficulty shifting their weight while sitting. This study assessed the immediate effects of circular trunk movements on the maximum reach distance and the centre of pressure distance that could be achieved during a multidirectional reach task in participants with post-stroke hemiparesis and healthy individuals as a control group. Methods This non-randomised controlled study included 26 participants with hemiparesis and 10 age-matched control participants. A functional reach apparatus, body pressure distribution measuring system and digital video camera were used to measure multidirectional reach, centre of pressure trajectories and circular trunk movements. The participants completed the assessments in the following order: multidirectional reach, circular trunk movements and repeat multidirectional reach. The authors classified the centre of pressure trajectories into type A (oval) and type B (non-oval) tracings according to the shape displayed by the circular trunk movements, which corresponded to the level of trunk function. The reach distance and centre of pressure distance before and after the circular trunk movements were compared. Results The reach distance and centre of pressure distance values in the front direction of participants with hemiparesis who exhibited type A tracings and had higher trunk function were significantly greater after completing circular trunk movements. In contrast, the reach distance and centre of pressure distance values of participants in all directions who exhibited type B tracings and had lower trunk function were significantly greater after completing circular trunk movements. Conclusions Circular trunk movements appeared to affect both reach distance and centre of pressure distance, particularly in participants with hemiparesis with type B tracings and lower trunk function. Implications for practice Physical and occupational therapists could incorporate circular trunk movements into rehabilitation programmes to help individuals with post-stroke hemiparesis improve their reach distance, trunk function and ability to shift their weight while sitting.

Assessing the data quality of AdHawk MindLink eye-tracking glasses.

Most commercially available eye-tracking devices rely on video cameras and image processing algorithms to track gaze. Despite this, emerging technologies are entering the field, making high-speed, cameraless eye-tracking more accessible. In this study, a series of tests were conducted to compare the data quality of MEMS-based eye-tracking glasses (AdHawk MindLink) with three widely used camera-based eye-tracking devices (EyeLink Portable Duo, Tobii Pro Glasses 2, and SMI Eye Tracking Glasses 2). The data quality measures assessed in these tests included accuracy, precision, data loss, and system latency. The results suggest that, overall, the data quality of the eye-tracking glasses was lower compared to that of a desktop EyeLink Portable Duo eye-tracker. Among the eye-tracking glasses, the accuracy and precision of the MindLink eye-tracking glasses were either higher or on par with those of Tobii Pro Glasses 2 and SMI Eye Tracking Glasses 2. The system latency of MindLink was approximately 9 ms, significantly lower than that of camera-based eye-tracking devices found in VR goggles. These results suggest that the MindLink eye-tracking glasses show promise for research applications where high sampling rates and low latency are preferred.

Three-Stage Deep Learning Framework for Video Surveillance

The escalating use of security cameras has resulted in a surge in images requiring analysis, a task hindered by the inefficiency and error-prone nature of manual monitoring. In response, this study delves into the domain of anomaly detection in CCTV security footage, addressing challenges previously encountered in analyzing videos with complex or dynamic backgrounds and long sequences. We introduce a three-stage deep learning architecture designed to detect abnormalities in security camera videos. The first stage employs a pre-trained convolutional neural network to extract features from individual video frames. Subsequently, these features are transformed into time series data in the second stage, utilizing a blend of bidirectional long short-term memory and multi-head attention to analyze short-term frame relationships. The final stage leverages relative positional embeddings and a custom Transformer encoder to interpret long-range frame relationships and identify anomalies. Tested on various open datasets, particularly those with complex backgrounds and extended sequences, our method demonstrates enhanced accuracy and efficiency in video analysis. This approach not only improves current security camera analysis but also shows potential for diverse application settings, signifying a significant advancement in the evolution of security camera monitoring and analysis technologies.

ЭКСПЕРИМЕНТАЛЬНАЯ ОЦЕНКА ТОЧНОСТИ ПРОСТРАНСТВЕННОЙ СТАБИЛИЗАЦИИ КВАДРОКОПТЕРА DJI AIR 2S*

The paper proposes a method for assessing the actual accuracy of spatial stabilization of a DJI Air 2S quadcopter using its video camera and processing the resulting video sequences with a developed software package in the mode of hovering over a reference object. The software package was implemented using the OpenCV library in Python. To register deviations of stabilization systems from a given position, it is proposed to use a reference field with known geometric parameters, information about which is entered into the processing algorithm. The main task of the software package is to identify the boundaries of the reference field and measure the geometric sides in pixels which allows you to determine the center of the field and measure its deviation relative to the center of the image as well as determine the height of the unmanned aerial vehicle flight. In the work the height measurement algorithm was tested. This technique is used to evaluate the accuracy of quadcopter stabilization with various combinations of navigation sensors used for this purpose. A statistical analysis of the obtained results was carried out and conclusions were drawn about compliance with the characteristics of positioning accuracy declared by the manufacturer using optical obstacle avoidance systems and satellite navigation systems.

Real time water surface elevation using video camera

In recent years, image processing has become a popular tool in wave flume experiments. In Indonesia, the use of image processing methods for measuring instruments in wave flume experiments has not been widely used. Therefore, in this study, an image processing method using image data from a digital camera was used in real time to measure the water surface elevation in the wave flume. Regular and irregular waves were measured with this image processing technique. The measurement results using this video camera were calibrated with wave gauges. The results showed that the measurement method using video images has a small difference less than 7% in wave height and wave period error. Therefore, this method can be a solution as a low cost measuring tool in wave flume or in field observations.

Software system for contactless text input based on computer vision

Abstract. The article is devoted to the problem of contactless human-computer interaction. The task of entering text from an optimized on-screen keyboard using cursor control based on data received from a video camera is considered, including if the user has motor impairments. A software system consisting of two applications is proposed for cursor control and text input based on computer vision algorithms. A diagram of the system operation and interaction of components and basic user settings are presented. As a result, testing was carried out using a focus group, in which participants had to enter free text using the proposed software system. Experiments have shown the functionality of the system.

Anatomy of stem lesions caused by Citrus leprosis

Abstract The citrus leprosis is responsible for causing major damage in fruit production. The disease, in the stems, is characterized by caused “scaly bark” lesions of reddish tone, depending on the stage they can be called blisters and make ruptures in the longitudinal direction. The aim of this study was to analyze the evolution of leprosis lesions of type CiLV-C and the anatomical alterations in the twigs of field-collected plants in order to compare to the evolution of the foliar and twigs lesions of inoculated plants already described in literature. For anatomical analysis, the samples were fixed in Karnovsky solution, dehydrated in a graded ethylic series, embedded in hidroxy-ethyl methacrylate resin (Leica Historesin), sectioned (5-7 μm thick), and stained with Toluidine blue for usual histological analysis. Histochemical tests were also carried out. The digital images were capture in a microscope with video camera. The analyzed lesion followed the same pattern as described to foliar lesions, i.e., there was a necrotic center surrounded by a halo also necrotic. In the necrotic region there was the accumulation of lipid compounds. The cortical and, or phloematic parenchyma exhibited hyperplasia. In both lesions it was verified the decrease of starch amount and the increase of quantity of prismatic calcium oxalate crystals, and the presence of gummosis traumatic ducts.

The effect of a hiding space on the behaviour and heart rate variability of dairy calves during temporary separation from the dam.

In natural settings, newborn calves hide for several days before joining the herd. It is unclear whether dairy calves housed indoors would show similar hiding behaviour. This study aimed to describe the use of an artificial hide provided to calves during temporary separation from the dam and assess the effect it has on lying and sleep-like behaviour, as well as heart rate variability (HRV). Twenty-eight cow-calf pairs were randomly assigned to having a hide (n = 14), or no hide (n = 14). Hide use (n = 14), as well as lying and sleep-like behaviour (n = 28), were recorded continuously via video camera during the first hour after the dam was removed for morning milking on day three to seven. Heart rate and R-R intervals were recorded using Polar equine monitors for a subsample of 12 calves (n = 6 per treatment) on day six. Descriptive statistics were calculated for hide use. Wilcoxon Signed Rank tests were used to evaluate whether having a hide affected lying and sleep-like behaviours as well as HRV. Hide use decreased over days and was highly variable between calves. Lying behaviour did not differ between treatments. Duration of sleep-like behaviour was higher for calves without a hide compared to those with a hide. Calves with a hide tended to show signs of higher HRV and parasympathetic activity compared to calves without a hide. Results suggest that providing a hiding space to young calves may be beneficial during periods when the cow is removed from the pen for milking.