Background: Telemedicine video consultations are rapidly increasing globally, accelerated by the COVID- 19 pandemic. This presents opportunities to use computer vision technologies to augment clinician visual judgement because video cameras are so ubiquitous in personal devices and new techniques, such as DeepLabCut (DLC) can precisely measure human movement from smartphone videos. However, the accuracy of DLC to track human movements in videos obtained from laptop cameras, which have a much lower FPS, has never been investigated; this is a critical gap because patients use laptops for most telemedicine consultations. Objectives: To determine the validity and reliability of DLC applied to laptop videos to measure finger tapping, a validated test of human movement. Method: Sixteen adults completed finger-tapping tests at 0.5 Hz, 1 Hz, 2 Hz, 3 Hz and at maximal speed. Hand movements were recorded simultaneously by a laptop camera at 30 frames per second (FPS) and by Optotrak, a 3D motion analysis system at 250 FPS. Eight DLC neural network architectures (ResNet50, ResNet101, ResNet152, MobileNetV1, MobileNetV2, EfficientNetB0, EfficientNetB3, EfficientNetB6) were applied to the laptop video and extracted movement features were compared to the ground truth Optotrak motion tracking. Results: Over 96% (529/552) of DLC measures were within +∕−0.5 Hz of the Optotrak measures. At tapping frequencies >4 Hz, there was progressive decline in accuracy, attributed to motion blur associated with the laptop camera’s low FPS. Computer vision methods hold potential for moving us towards intelligent telemedicine by providing human movement analysis during consultations. However, further developments are required to accurately measure the fastest movements.
keywords: پزشکی از راه دور | ضربه زدن با انگشت | موتور کنترل | کامپیوتری | Telemedicine | DeepLabCut | Finger tapping | Motor control | Computer vision

The combustion and fluidization behavior of biomass depend on the physical properties (size, morphology, and density) and mechanical performances (elastic modulus, Poisson’s ratio, tensile strength and failure strain), but their quantitative models have rarely been focused in previous researchers. Hence, a static image measurement for particle physical properties is studied. Combining the uniaxial tension and digital image correlation tech- nology, the dynamic image measurement method for the mechanical properties is proposed. The results indicate that the average roundness, rectangularity, and sphericity of present biomass particles are 0.2, 0.4, and 0.16, respectively. The equivalent diameter and density obey the skewed normal distribution. The tensile strength and failure stress are sensitive to stretching rate, fiber size and orientation. The distribution intervals of elastic modulus and Poisson’s ratio are 30–600 MPa and 0.25–0.307, respectively. The stress–strain curves obtained from imaging experiments agree well with the result of finite element method. This study provides the operating parameters for the numerical simulation of particles in the fluidized bed and combustor. Furthermore, the computer vision measurement method can be extended to the investigations of fossil fuels.
keywords: ذرات زیست توده | مشخصات فیزیکی | اجرای مکانیکی | تست کشش | آزمایش تصویربرداری | بینایی کامپیوتر | Biomass particle | Physical properties | Mechanical performances | Tensile testing | Imaging experiment | Computer vision

Some researchers using traditional taphonomic criteria (groove shape and presence/absence of microstriations) have cast some doubts about the potential equifinality presented by crocodile tooth marks and stone tool butchery cut marks. Other researchers have argued that multivariate methods can efficiently separate both types of marks. Differentiating both taphonomic agents is crucial for determining the earliest evidence of carcass processing by hominins. Here, we use an updated machine learning approach (discarding artificially bootstrapping the original imbalanced samples) to show that microscopic features shaped as categorical variables, corresponding to intrinsic properties of mark structure, can accurately discriminate both types of bone modifications. We also implement new deep-learning methods that objectively achieve the highest accuracy in differentiating cut marks from crocodile tooth scores (99% of testing sets). The present study shows that there are precise ways of differentiating both taphonomic agents, and this invites taphonomists to apply them to controversial paleontological and archaeological specimens.
keywords: تافونومی | علائم برش | علائم دندان | فراگیری ماشین | یادگیری عمیق | شبکه های عصبی کانولوشنال | قصابی | Taphonomy | Cut marks | Tooth marks | Machine learning | Deep learning | Convolutional neural networks | Butchery

The effect of surfactants on vertical gas-liquid flow is experimentally investigated in a 12.7 mm diameter tube at conditions relevant to an ammonia-water bubble absorber. The characteristics of two-phase flow are studied using an air-water mixture, both with and without the addition of 1-octanol as the surfac- tant. High-speed videography is used to study the flow patterns and quantify interfacial areas and bubble velocities. Novel computer vision-based methods are used to analyze and quantify these flow parame- ters. The addition of 1-octanol results in enhancement in interfacial area due to the prevention of bubble coalescence leading to many small diameter bubbles. Measured values of interfacial area are compared with predictions from correlations in the literature, and agreement and differences are interpreted and discussed. The bubble velocity is measured by object tracking using the optical flow method. Surfactants lead to a decrease in bubble velocity and increase in the residence time. These are surmised to be due to the shear stresses caused by the non-uniform concentration distribution of surfactant along the bub- ble surface. Overall, the addition of surfactants can lead to appreciable enhancement in heat and mass transfer rates due to their effect on interfacial areas and residence times.
keywords: سورفکتانت ها | جریان دو فازی | ناحیه رابط | سرعت | تقویت | تجسم جریان | Surfactants | Two-phase flow | Interfacial area | Velocity | Enhancement | Flow visualization

Using polar stereographic images from the Mars Color Imager (MARCI), we use Python to autonomously track the Northern Polar Seasonal Cap (NPSC) recession from Mars Years (MY) 29 to MY 35 between Ls = 10° and Ls = 70°. We outline the cap and find an ellipse of best fit. We then compare our results to previously published recession rates, that were manually tracked, and find them to be consistent. Our process benefits from being automated, which increases the speed of tracking and allows us to monitor the recession with higher Ls fidelity than past studies. We find that most MYs have a local minimum recession rate at Ls = ~32° and a local maximum at Ls = ~51°. We also find that MY 30 experiences a rapid latitude increasing event that involves ~1° Ls of a rapid increase and ~5° Ls of slower recession, which then increases above the interannual average rate. We interpret this to be the result of a major sublimation driven by off-polar winds. We also discover divergent effects in the recession and size of the NPSC following the MY 28 and MY 35 global dust storms. MY 29’s cap is significantly smaller and retreats slower than the multi-year average, whereas MY 35’s cap is slighter larger and retreats very close to the average. We hypothesize that the diverging behavior of the caps in post-storm years can be a result of the differences in the date of onset and the duration of the storms.