Video Camera Research Articles

G-GOP: Generative Pose Estimation of Reflective Texture-Less Metal Parts With Global-Observation-Point Priors

6-D pose estimation of reflective and texture-less metal parts remains a challenge in intelligent manufacturing. A feature-to-image method has been recently proposed to address this problem. However, this method still contains unclear factors that impact its precision and robustness, as well as those of other conventional methods. To address this issue, first, the mechanisms of conventional methods are theoretically analyzed, and we find and summarize two factors. The errors caused by these factors are defined as the discrete template substitution (DTS) error, which is caused by the discretization transformation between intermediate results and final poses, and the center approximation substitution (CAS) error, which is caused by differences in contour projections in different local fields of view. Moreover, a novel feature-to-image method is proposed to reduce these errors. It utilizes a novel pose representation called the observation pose. Unlike the traditional pose, it is represented with six distinct camera motions to decouple the influence of different parameters on the projected image. The conversion between it and the traditional pose is not discrete, so it can reduce the DTS error. Additionally, a new prior called GOP is used, which is a global point and converted from the output of the detector to describe the local field of view. Compared to conventional methods, the proposed method can establish a mapping between different local fields of view and differences in contour projection to reduce CAS error. Three experiments were carried out and compared with different methods. Experiments contain 2250 different RGB images captured by a single MV-CA050-11UC industrial RGB camera fixed on the MELFA RV13FD 6-DoF robot. Compared with state-of-the-art methods, the ADD-S, average rotation, and translation precision were increased by 4.8%, 23%, and 17%, respectively, which demonstrate that the proposed method has superior robustness and precision.

Deep squat test – Functional movement Screen: Convergent validity and ability to discriminate subjects with different levels of joint mobility

BackgroundFunctional Movement Screen (FMS) is an important tool in the assessment of exercise practice. Assuming FMS lacks precise validity for assessing postural deficits, further research is needed to assess whether it is a sufficiently precise tool for analysing joint mobility. Research aims were to evaluate: convergent validity of Deep Squat (DS) – one of FMS tests – regarding joint mobility, using data from a three-dimensional motion analysis as a comparable method; DS's ability to discriminate between subjects with different joint mobility levels. MethodsSixty subjects were selected (23.6 ± 3.8 years). DS was performed according to FMS guidelines. Subjects' performance in frontal and sagittal planes was recorded by two video cameras and subsequently scored by two FMS-certified evaluators. Three-dimensional motion analyses of DS were acquired by a Vicon Motion Capture System (200 Hz). Ten trials were acquired for each subject. Ankle, knee, hip, and shoulder angular positions in sagittal plane were determined from the FullBody PlugInGait model. Spearman's coefficient examined the correlation between angular positions and DS score. Kruskal-Wallis test was used to assess the DS ability to discriminate between subjects with different joint mobility levels by comparing different scores. ResultsNegligible to moderate correlations were found between DS score and angular positions (−0.5 < r < 0.5). Only shoulder angular positions showed differences between score “1” and “2” (p < 0.05). Shoulder and hip angular positions showed no differences between score “2” and “3” (p < 0.05). ConclusionsDS yielded low convergent validity regarding joint mobility and did not show the ability to discriminate between subjects with different joint mobility levels.

The Influence of Pedestrian Characteristic on Pedestrian Speed in Urban Area: A Case Study at Jalan Tuanku Abdul Rahman, Kuala Lumpur

Short-distance walking, alone or with additional modes, is efficient. Well-established public transport may increase the number of pedestrians since they prefer walking to work, travel, and shop because cities are connected by activities and locations. However, pedestrian-friendly paths and public transport in Malaysia have received little attention. Large-scale facilities like stadiums, movie theatres, and amusement parks are driving the need for pedestrian facilities. Thus, this study aims to examines the pedestrian characteristic at Jalan Tuanku Abdul Rahman, Kuala Lumpur. To discover significant elements, pedestrian behaviour on sidewalks at the indicated location was quantitatively investigated, taking into account general pedestrian characteristics that affect walking speed, such as age, gender, and trip companion. Using a video camera, pedestrian variables like age, gender, and trip companion were calculated for volume, speed, flow, density, and space during weekday and weekend peak hours. Extracted data were then analysed by using SPSS as a statistical software and the Highway Capacity Manual (HCM 2010) was used to classify pedestrian level of services (LOS). Age, gender, and companionship substantially determine the pedestrian flow and the study’s LOS for pedestrian characteristics ranges from D to E, indicating that Jalan Tuanku Abdul Rahman’s sidewalk relationship varies. This research strongly recommends a dedicated walkway to increase pedestrian space and improve services along Jalan Tuanku Abdul Rahman, Kuala Lumpur. This study will also serve as a guideline to the transport operator into improving the future of pedestrian walkway design facilities.

Examining the interfacial behavior of non-Newtonian gas-liquid two-phase flow in horizontal square microchannels

This study investigates the flow dynamics of a non-Newtonian two-phase flow within a horizontal square microchannel. The experimental apparatus employed an acrylic channel with a side length of 8 × 10−4 m. Nitrogen served as the test gas, and the test liquids consisted of water and CMC (Carboxymethylcellulose) aqueous solutions at concentrations of 0.2 and 0.4 %wt. The superficial velocity of working gas and liquids spanned ranges of 0.26–20 m/s and 0.05–1.0 m/s, respectively. A differential pressure transducer was used to measure pressure gradients, while a high-speed video camera captured the flow behavior for analysis with a custom image processing technique. Nonlinear regression statistical analysis was applied to establish the criteria for flow pattern transition. Furthermore, signal processing techniques such as PSD (Power Spectral Density), DWT (Discrete Wavelet Transform), ANN (Artificial Neural Network), and Kolmogorov entropy were employed to analyze the flow behavior based on the pressure gradient signals from the differential pressure transducers. Our findings reveal the occurrence of bubbly, slug, slug-annular, and churn flow patterns. The nonlinear regression statistical analysis offered ambiguous results for determining flow pattern transitions. However, an empirical coefficient was obtained for the nondimensional liquid height. Moreover, the empirical nondimensional liquid height aligned well with the experimental values. The PSD and DWT analysis corroborated each other in clearly characterizing the flow pattern based on pressure gradient fluctuation and wavelet energy. The Kolmogorov entropy effectively captured the chaotic flow pattern. Furthermore, the ANN analysis demonstrated a solid agreement between the predicted and actual flow patterns.

On the performance of streamflow gauging using CCTV-integrated LSPIV in diverse hydro-environmental conditions.

Addressing the critical need for precise streamflow measurements in hydro-environmental research, this study evaluates large-scale particle image velocimetry (LSPIV) using cost-effective closed-circuit television (CCTV) cameras, providing a detailed sensitivity analysis in both laboratory and real-world canal settings. In laboratory conditions, a 45° camera angle notably enhanced performance, exhibiting a 12% decrease in MAE and a remarkable 40% reduction in RMSE compared to the performance of orthographic form. Tracer particles further enhanced LSPIV accuracy, decreasing both mean absolute error (MAE) and root mean square error (RMSE) by around 0.05 m/s. Optimal velocity coefficients for the lab ranged between 0.85 and 0.90. Nighttime measurements, using projection-based illumination, showed a minor 3% MAE variation and 0.02 RMSE difference versus daytime. In field experiments, a 45° upstream CCTV camera configuration notably improved LSPIV accuracy, achieving a 3% MAE and 0.055 m/s RMSE. For best results across different turbidity levels, we recommend a velocity coefficient range of 0.84 to 0.88. This study highlights the robustness and cost-efficiency of LSPIV as a transformative method for streamflow gauging, demonstrating its wide applicability in diverse hydro-environmental scenarios.

Experimental study of the time of pressure rise and fall in the sprayer pipe

A wide variety of meteorological factors, soil and climatic conditions, the saturation of fields with many types of weeds, a significant set of cultivated plants and many other factors necessitate the implementation of innovative technological schemes for the use of pesticides, which will reduce the pesticide load as much as possible and determine the safe environmental effect of preparations. Experimental studies were carried out to determine the time of pressure rise and fall in the sprayer pipe and to establish the corresponding functional dependencies. The automatic adjustment system of the application rate was set to change the pressure when the green simulation surface hit 0.5 MPa, and the yellow surface hit 0.2 MPa. Sensor impact on a surface of different colours was simulated. At the same time, the pressure in the system changed. The pressure in the sprayer pipe of trailer sprayer OP-2000-2 was measured using a WIKAI EN837-1 pressure gauge and the readings were recorded with a Panasonic NV-GS75 digital video camera mounted on a Continent TRIPOD A2 stand. As a result of calculating the experimental data, it was found that the arithmetic mean of pressure reduction from 0.5 to 0.2 MPa is 1.94 s. And the arithmetic mean for four experiments, a pressure reduction of 0.05 MPa is 0.32 s. When calculating the experimental data of the pressure rise, it was found that the arithmetic mean of the pressure rise from 0.2 to 0.5 MPa is 2.37 s. And the average value for four experiments, the pressure rise by 0.05 MPa is 0.395 s. After approximation of experimental data obtained during single-factor experiments, when the pressure P and the corresponding time values t change, the corresponding polynomial functions describing the characteristic t=f(P) when operating a sprayer with automatic application rate control were obtained. The results obtained would allow calculating the time of delay or advance opening of the spray torch when working with variable standards for target plans, in order to avoid non-compliance with the application rate with the specified values

A 6-DOF camera motion correction method using IMU sensors for photogrammetry and optical measurements

Environmental conditions such as wind and ground traffic introduce motion in camera measurement systems and affect measurement accuracy. Conventional camera motion correction methods track static reference points with one or multiple cameras, reducing applicability. This study proposes a novel 6-degree-of-freedom (DOF) camera motion correction method using only an inertial measurement unit (IMU) sensor. A Kalman filter is adopted as a data fusion method to estimate the camera orientation and translation using IMU data. Six pinhole camera models are built to evaluate and correct 6-DOF camera motions. The motion correction efficiency and robustness are tested for different object distances and focal lengths of optical lenses. The motion correction ratio is statistically analysed and reaches approximately 80%. The object distance has little effect on the motion correction ratio. The rotation-induced pixel movement is independent of the object distance. More than 90% of the pixel movement noise is caused by camera rotation. The translation-induced pixel movement is inversely correlated with the object distance.

SELECTION AND JUSTIFICATION OF THE LOCATION OF CAMERAS OF THE ALLROUND VISION SYSTEM ON AN UNMANNED DUMP TRUCK

The choice of equipment installation location on the sides of an unmanned vehicle, and especially such a dump truck, is a difficult task, which is solved, as a rule, intuitively and by the method of fullscale modeling. The authors of the article have solved the formulated problem, on substantiation of the choice of places of sensorics arrangement on the dump truck, using the developed specialized software application. The Carla simulator was chosen as a tool for carrying out the research as it has wide possibilities of visualization of the surrounding world. At the initial stage of selecting the location of the camera arrangement, the requirements for their viewing angles and the quality of the image obtained from them were formulated. This required the development of a software application in the Python language using the Open3D library, the choice of which is justified by the fact that the Carla simulator provides access to its API through a library for this language, and also because it allows rapid prototyping of applications. The developed software application helps the user to select the location of the video cameras on the dump truck in order to experience the unmanned control system. Further research is aimed at automating the selection of camera locations on board an unmanned dump truck.

A marker-less human motion analysis system for motion-based biomarker identification and quantification in knee disorders.

In recent years the healthcare industry has had increased difficulty seeing all low-risk patients, including but not limited to suspected osteoarthritis (OA) patients. To help address the increased waiting lists and shortages of staff, we propose a novel method of automated biomarker identification and quantification for the monitoring of treatment or disease progression through the analysis of clinical motion data captured from a standard RGB video camera. The proposed method allows for the measurement of biomechanics information and analysis of their clinical significance, in both a cheap and sensitive alternative to the traditional motion capture techniques. These methods and results validate the capabilities of standard RGB cameras in clinical environments to capture clinically relevant motion data. Our method focuses on generating 3D human shape and pose from 2D video data via adversarial training in a deep neural network with a self-attention mechanism to encode both spatial and temporal information. Biomarker identification using Principal Component Analysis (PCA) allows the production of representative features from motion data and uses these to generate a clinical report automatically. These new biomarkers can then be used to assess the success of treatment and track the progress of rehabilitation or to monitor the progression of the disease. These methods have been validated with a small clinical study, by administering a local anaesthetic to a small population with knee pain, this allows these new representative biomarkers to be validated as statistically significant (-value ). These significant biomarkers include the cumulative acceleration of elbow flexion/extension in a sit-to-stand, as well as the smoothness of the knee and elbow flexion/extension in both a squat and sit-to-stand.

Advantages and Indications for Laparoscopic Cholecystectomy: A Review

Laparoscopic cholecystectomy quickly became the procedure of choice for treating patients with biliary discomfort and gallstone problems after the initial reports appeared in the late 1980s. Technical advancements, particularly in miniature video cameras and specialized equipment, played a crucial role in the evolution of laparoscopic cholecystectomy. Subsequent improvements in the design and production of instruments and equipment have further enhanced the safety and efficacy of minimally invasive surgery, extending its application to the treatment of various gastrointestinal illnesses, including gallstone disease.

Swelling of rubbers of different chemical natures in supercritical carbon dioxide

Objectives. To investigate the swelling of the main types of rubbers used in the rubber industry in carbon dioxide in a supercritical state (SC-CO2), in order to assess the possibility of obtaining elastomeric materials with porous structures using fluid technology, based on them.Methods. The process of swelling of rubbers in SC-CO2 and subsequent foaming was carried out according to a specially developed technique using the original installation. This is a high-pressure apparatus with transparent windows, allowing for the use of an optical technique to directly measure the geometric dimensions of samples during swelling and foaming using a digital video camera. The study of the porous structure of foamed rubbers was carried out using scanning electron microscopy.Results. The study established experimental curves of the swelling kinetics in SC-CO2 of isoprene, butadiene, styrene butadiene, ethylene propylene, chloroprene, ethylene acrylate, siloxane, and organofluorine rubbers. The influence of temperature and pressure on the rate and equilibrium degree of swelling was studied. The diffusion coefficients of SC-CO2 in rubbers of various chemical natures were also determined.Conclusions. It was shown that the equilibrium swelling degree of rubbers in SC-CO2 depends on the chemical nature of rubbers. It does not correlate with the value of their solubility parameters, changes directly proportional to the diffusion coefficient and increases with increasing temperature and pressure. It was found that irrespective of the degree of swelling in SC-CO2, all the rubbers studied are intensively foamed at a sharp pressure drop. The size of the pores formed is tens of microns: significantly smaller than the size of pores formed when chemical pore formers are used.

Effectiveness of a dispatcher-assisted CPR using an animated image: Simulation study

IntroductionConsidering the few studies evaluating bystander cardiopulmonary resuscitation (CPR) performance, we sought to analyze differences in bystander CPR performance with and without the use of our self-developed animated GIFs based on dispatcher-assisted CPR simulation. MethodsA total of 80 adults who had not received CPR training over the past two years participated in the study. Among them, 40 people were classified into the auditory group (receiving CPR instructions only over the phone), and the other 40 people were classified into the audiovisual group (receiving CPR instructions over the phone after receiving images on a smartphone). All participants were asked to perform adult and infant CPR for 2 min. CPR performance was recorded using two video cameras (front and side) and analyzed by two emergency physicians, whereas CPR quality was measured using Resusci Anne & Baby QCPR Mk II (Laerdal). ResultsIn the adult CPR study, the audiovisual group had higher performance scores for adequacy of “knee position,” “hand posture,” “elbow extension,” and “vertical compression,” as well as higher Standard Posture Completeness and Instruction Performance scores (p < 0.001). No significant difference in CPR quality was observed between the two groups. In the infant CPR study, audiovisual group had higher performance scores in adequacy of “compression site,” “finger posture,” and “vertical compression,” as well as higher Standard Posture Completeness and Instruction Performance scores (p < 0.001). Regarding CPR quality, the audiovisual group had higher scores for “adequate compression rate ratio” (p = 0.047). ConclusionAudiovisual guidance using animated GIFs more effectively improved CPR Standard Posture Completeness and Instruction Performance than did traditional auditory guidance.

The Disintegration of Magnetic Fluid Layer on Liquid and Solid Substrates in Vertical Magnetic Field

The purpose of this work is to experimentally study the process of decomposition of a horizontal layer of a magnetic fluid lying on a liquid non-wettable substrate into an ordered system of drops under the influence of a vertical magnetic field.Methods. In the experiment, the studied liquids in the form of a two-layer system filling a cylindrical glass cuvette were placed on a horizontal platform in the center of the Helmholtz coils connected to a direct current source. The development of magnetic fluid layer free surface instability was monitored using a high-speed digital video camera installed at the top of the coil system.Results. The dependence of the critical strength of the magnetic field on the thickness of the torn layer and the magnetic susceptibility of the magnetic fluid is determined. The obtained experimental data are compared with the results of existing theoretical studies of the instability of the magnetic fluid layer. Compared to the case of a solid substrate, the rupture of the magnetic fluid layer on a liquid substrate occurred at lower values of the critical field strength. In this case, the development of instability of the free surface magnetic fluid layer on a solid substrate occurs at field strengths twice as high as the instability strength of the interface of the layer on a liquid substrate.Conclusion. The use of a liquid substrate makes it possible to break such layers of the magnetic fluid, the deformation of which on a solid substrate is limited only by a periodic perturbation of the surface. The magnitude of the critical strength leading to the disintegration of a continuous magnetic fluid layer on a liquid substrate increases with increasing thickness of this layer. An increase in the magnetic susceptibility of the magnetic fluid leads to a decrease in the critical field strength.

MV-MS-FETE: Multi-view multi-scale feature extractor and transformer encoder for stenosis recognition in echocardiograms

Background: aortic stenosis is a common heart valve disease that mainly affects older people in developed countries. Its early detection is crucial to prevent the irreversible disease progression and, eventually, death. A typical screening technique to detect stenosis uses echocardiograms; however, variations introduced by other tissues, camera movements, and uneven lighting can hamper the visual inspection, leading to misdiagnosis. To address these issues, effective solutions involve employing deep learning algorithms to assist clinicians in detecting and classifying stenosis by developing models that can predict this pathology from single heart views. Although promising, the visual information conveyed by a single image may not be sufficient for an accurate diagnosis, especially when using an automatic system; thus, this indicates that different solutions should be explored.Methodology: following this rationale, this paper proposes a novel deep learning architecture, composed of a multi-view, multi-scale feature extractor, and a transformer encoder (MV-MS-FETE) to predict stenosis from parasternal long and short-axis views. In particular, starting from the latter, the designed model extracts relevant features at multiple scales along its feature extractor component and takes advantage of a transformer encoder to perform the final classification.Results: experiments were performed on the recently released Tufts medical echocardiogram public dataset, which comprises 27,788 images split into training, validation, and test sets. Due to the recent release of this collection, tests were also conducted on several state-of-the-art models to create multi-view and single-view benchmarks. For all models, standard classification metrics were computed (e.g., precision, F1-score). The obtained results show that the proposed approach outperforms other multi-view methods in terms of accuracy and F1-score and has more stable performance throughout the training procedure. Furthermore, the experiments also highlight that multi-view methods generally perform better than their single-view counterparts.Conclusion: this paper introduces a novel multi-view and multi-scale model for aortic stenosis recognition, as well as three benchmarks to evaluate it, effectively providing multi-view and single-view comparisons that fully highlight the model's effectiveness in aiding clinicians in performing diagnoses while also producing several baselines for the aortic stenosis recognition task.

Enabling Personalization for Digital Cognitive Stimulation to Support Communication With People With Dementia: Pilot Intervention Study as a Prelude to AI Development.

Maintaining good communication and engagement between people with dementia and their caregivers is a major challenge in dementia care. Cognitive stimulation is a psychosocial intervention that supports communication and engagement, and several digital applications for cognitive stimulation have been developed. Personalization is an important factor for obtaining sustainable benefits, but the time and effort required to personalize and optimize applications often makes them difficult for routine use by nonspecialist caregivers and families. Although artificial intelligence (AI) has great potential to support automation of the personalization process, its use is largely unexplored because of the lack of suitable data from which to develop and train machine learning models. This pilot study aims to evaluate a digital application called Aikomi in Japanese care homes for its potential to (1) create and deliver personalized cognitive stimulation programs to promote communication and engagement between people with dementia and usual care staff and (2) capture meaningful personalized data suitable for the development of AI systems. A modular technology platform was developed and used to create personalized programs for 15 people with dementia living in 4 residential care facilities in Japan with the cooperation of a family member or care staff. A single intervention with the program was conducted with the person with dementia together with a care staff member, and for some participants, smell stimulation was provided using selected smell sticks in conjunction with the digital program. All sessions were recorded using a video camera, and the combined personalized data obtained by the platform were analyzed. Most people with dementia (10/15, 67%) showed high levels of engagement (>40 on Engagement of a Person with Dementia Scale), and there were no incidences of negative reactions toward the programs. Care staff reported that some participants showed extended concentration and spontaneous communication while using Aikomi, which was not their usual behavior. Smell stimulation promoted engagement for some participants even when they were unable to identify the smell. No changes in well-being were observed following the intervention according to the Mental Function Impairment Scale. The level of response to each type of content in the stimulation program varied greatly according to the person with dementia, and personalized data captured by the Aikomi platform enabled understanding of correlations between stimulation content and responses for each participant. This study suggests that the Aikomi digital application is acceptable for use by persons with dementia and care staff and may have the potential to promote communication and engagement. The platform captures personalized data, which can provide suitable input for machine learning. Further investigation of Aikomi will be conducted to develop AI systems and create personalized digital cognitive stimulation applications that can be easily used by nonspecialist caregivers.

Inferring benthic megafaunal sediment reworking activity in relation to bottom water oxygen in Barkley Canyon, NE Pacific from video and acoustic imaging analysis

Sediment reworking activity influences benthic functioning expressed as nutrient fluxes and carbon cycling. Multiple studies have addressed sediment reworking based on observations from individual instrument types (e.g., video camera, side-scan sonar, multibeam), but none have considered reworking based on two or more complementary instruments. We therefore analyzed deep-sea megafaunal and reworked sediment traces by combining and comparing observations from non-invasive instruments, an underwater video camera and a rotary sonar, as part of the node of Ocean Networks Canada's NEPTUNE cabled observatory, located at 396 m depth in Barkley Canyon Upper Slope, NE Pacific Ocean. Specifically, we examined sediment traces and benthic megafaunal communities during two different sampling periods (May and September 2013) and at two different spatial scales of analysis. The camera images (∼0.5 m2) documented significantly lower megafaunal density during the period of reduced oxygen concentrations (May). Although we did not observe significant differences in sediment trace diversity and density between the two study periods, we were interested in how the relief traces generated by unidentified gastropods (likely the family Solariellidae) influenced sediment mixing. Relief traces showed higher density in low oxygen (May, 1.87 count m−2) than in high oxygen (September, 1.17 count m−2) conditions. Sonar images (∼1268 m2), which lacked sufficient resolution to allow identification of benthic organisms, documented distributions of biological pits (regions of low backscatter), a significantly greater proportion of bioturbated seafloor area, and increased in pit size with increased oxygen levels. Pits dominated sonar images at this location and persisted through the two study periods. Most sonar field of view subsections showed a significant increase in circularity of pit shapes, likely explained by increased reworked sediment with increasing oxygen concentration. We conclude that, except for relief burrows, higher reworked sediment area coincided with the highest oxygen concentration, which aligns with previously established reduced metabolic activity by benthos adapted to oxygen minimum zones. Furthermore, we emphasize the complementarity of the two imaging techniques (video and sonar) in understanding deep-sea benthic ecosystem dynamics, as well as the importance of considering multiple spatial scales when investigating proxies of bioturbation activity.

BusEye: A multi-stream approach for driver behavior analysis on public bus driver cameras

In the contemporary era, modern public buses are equipped with multiple built-in cameras, offering real-time or recorded monitoring capabilities through centralized systems. This research endeavors to introduce a novel approach for classifying driver behaviors at the signalized road intersections with traffic lights by leveraging a public bus vehicular camera that encompasses the road and thus traffic lights. The proposed method involves a 2-step multi-head self-attention network that employs the extracted traffic light state, and optic flow information besides the raw vehicular camera video. Subsequently, a bi-directional long short-term memory based spatio-temporal aggregation strategy is applied to guide the learning final discriminative representation. To evaluate the proposed approach, we introduce a benchmark dataset called BusEye, which covers the challenging real-life scenarios that are captured on actual vehicular cameras mounted on public busses. We provide extensive analysis and ablation studies on the BusEye dataset that shows our proposed solution outperforms the state-of-the-art video classification methods both on a per-class basis and overall. Based on the outcomes derived from the validation and test data, the proposed method demonstrated remarkable efficacy in classifying driver behaviors at signalized intersections.

Computer Vision-Based Pier Settlement Displacement Measurement of a Multispan Continuous Concrete Highway Bridge under Complex Construction Environments

Various concrete bridges have been built across oceans, valleys, and mountains; however, the settlement displacement of bridge piers caused by environmental changes or self-weight during construction phases often leads to uneven stresses, cracking, and eventual collapse. To address the labor-intensive and high-cost issues of pier displacement monitoring using contact-type sensors, this paper proposes an automatic vision-based method for measuring pier settlement displacement under complex construction environments, such as complex image backgrounds, varying ambient light, and camera movement. In the proposed method, a deep learning network was first employed to eliminate the adverse effect of complex image backgrounds and varying ambient light on the accuracy of target detection; then, an adaptive displacement extraction algorithm without a human-computer interaction process was developed to automatically extract the center coordinates of targets attaching to the bridge piers and reference platform; finally, the pier settlement displacement was calculated by using the relative displacements obtained by a dual camera system to eliminate the measurement error caused by camera translation and rotation movements. Laboratory tests of a cantilever beam and field tests of a continuous multispan concrete girder highway bridge under construction have successfully validated the effectiveness and robustness of the developed methodology. The results obtained in this paper can provide some insights for engineers in applying computer vision technology for the real-time monitoring of bridge displacements.

Resistance of yielding rockbolts to multiple impact loads

The resistance of yielding rockbolts to multiple impact loads was tested by means of a drop hammer. The methodology was based on the ASTM D7401-08 standard as well as the requirements specified for rockbolts by Luossavaara-Kiirunavaara Aktiebolag (LKAB - a Swedish mining company) and the Safety In Mines Research Advisory Committee of South Africa (project number GAP 423). During the tests, the loading force and bolt elongation were measured with a sampling frequency of fs = 19.2 kHz. In order to analyse the phenomena involved, the tests were recorded using two independent high-speed video cameras (600 and 1000 frames per second) and one thermal camera (128 frames per second). The tests demonstrated that all the samples transferred double the gravitational potential energy of Ep = 50.85 kJ from the impact load of mass m = 2825 kg and impact velocity of v = 6.0 m/s without failure. Damage to the bolt-resin contact in the upper anchoring zone of the bolt in the steel tube occurred after subsequent impacts. As a result, the macro-deformed bolt exhibits additional yield, and its operation following the shearing of the resin bond is similar to that of a 'cone bolt'. Only one test resulted in damage to the threaded bolt rod-nut coupling, during the fourth impact; however, no damage to the bolt-resin contact zone was observed.

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.