Algorithms for Vision-Based Quality Control of Circularly Symmetric Components.

Quality inspection in the industrial production field is experiencing a strong technological development that benefits from the combination of vision-based techniques with artificial intelligence algorithms. This paper initially addresses the problem of defect identification for circularly symmetric mechanical components, characterized by the presence of periodic elements. In the specific case of knurled washers, we compare the performances of a standard algorithm for the analysis of grey-scale image with a Deep Learning (DL) approach. The standard algorithm is based on the extraction of pseudo-signals derived from the conversion of the grey scale image of concentric annuli. In the DL approach, the component inspection is shifted from the entire sample to specific areas repeated along the object profile where the defect may occur. The standard algorithm provides better results in terms of accuracy and computational time with respect to the DL approach. Nevertheless, DL reaches accuracy higher than 99% when performance is evaluated targeting the identification of damaged teeth. The possibility of extending the methods and the results to other circularly symmetrical components is analyzed and discussed.

Identification of Aluminothermic Reaction and Molten Aluminum Level through Vision System.

During the secondary production of aluminum, upon melting the scrap in a furnace, there is the possibility of developing an aluminothermic reaction, which produces oxides in the molten metal bath. Aluminum oxides must be identified and removed from the bath, as they modify the chemical composition and reduce the purity of the product. Furthermore, accurate measurement of molten aluminum level in a casting furnace is crucial to obtain an optimal liquid metal flow rate which influences the final product quality and process efficiency. This paper proposes methods for the identification of aluminothermic reactions and molten aluminum levels in aluminum furnaces. An RGB Camera was used to acquire video from the furnace interior, and computer vision algorithms were developed to identify the aluminothermic reaction and melt level. The algorithms were developed to process the image frames of video acquired from the furnace. Results showed that the proposed system allowed the online identification of the aluminothermic reaction and the molten aluminum level present inside the furnace at a computation time of 0.7 s and 0.4 s per frame, respectively. The advantages and limitations of the different algorithms are presented and discussed.

Effects of the whole-body vibration direction on the cognitive response of standing subjects.

This study aims to investigate the effect of whole-body vibration along different axes on the response time (RT) of standing subjects during a customised psychomotor vigilance task (PVT). Twenty-five subjects were exposed to harmonic vibration with amplitude of 0.7 m/s2 RMS and frequencies between 1.5 Hz and 12.5 Hz. ANOVA was used to assess if the difference of RT with and without vibration had a statistical relevance. Results showed that the RT was statistically affected by the vibration only at frequencies below 2 Hz. The vibration at higher frequencies had a minor effect on the RT. The RTs during the vibration exposure was, on average, 15% higher than the RT post exposure. Practitioner summary: This study investigates the effects of whole-body vibration (WBV) along different axes on the response time (RT). We measured the RTs to a psychomotor vigilance task of 25 standing subject exposed to WBV. The cognitive response was statistically affected by the WBV and, on average RT have increased of 15%.

Machine Learning-Based Estimation of Ground Reaction Forces and Knee Joint Kinetics from Inertial Sensors While Performing a Vertical Drop Jump.

Nowadays, the use of wearable inertial-based systems together with machine learning methods opens new pathways to assess athletes' performance. In this paper, we developed a neural network-based approach for the estimation of the Ground Reaction Forces (GRFs) and the three-dimensional knee joint moments during the first landing phase of the Vertical Drop Jump. Data were simultaneously recorded from three commercial inertial units and an optoelectronic system during the execution of 112 jumps performed by 11 healthy participants. Data were processed and sorted to obtain a time-matched dataset, and a non-linear autoregressive with external input neural network was implemented in Matlab. The network was trained through a train-test split technique, and performance was evaluated in terms of Root Mean Square Error (RMSE). The network was able to estimate the time course of GRFs and joint moments with a mean RMSE of 0.02 N/kg and 0.04 N·m/kg, respectively. Despite the comparatively restricted data set and slight boundary errors, the results supported the use of the developed method to estimate joint kinetics, opening a new perspective for the development of an in-field analysis method.

Machine-Learning Based Determination of Gait Events from Foot-Mounted Inertial Units.

A promising but still scarcely explored strategy for the estimation of gait parameters based on inertial sensors involves the adoption of machine learning techniques. However, existing approaches are reliable only for specific conditions, inertial measurements unit (IMU) placement on the body, protocols, or when combined with additional devices. In this paper, we tested an alternative gait-events estimation approach which is fully data-driven and does not rely on a priori models or assumptions. High-frequency (512 Hz) data from a commercial inertial unit were recorded during 500 steps performed by 40 healthy participants. Sensors' readings were synchronized with a reference ground reaction force system to determine initial/terminal contacts. Then, we extracted a set of features from windowed data labeled according to the reference. Two gray-box approaches were evaluated: (1) classifiers (decision trees) returning the presence of a gait event in each time window and (2) a classifier discriminating between stance and swing phases. Both outputs were submitted to a deterministic algorithm correcting spurious clusters of predictions. The stance vs. swing approach estimated the stride time duration with an average error lower than 20 ms and confidence bounds between ±50 ms. These figures are suitable to detect clinically meaningful differences across different populations.

Four degree-of-freedom lumped parameter model of the foot-ankle system exposed to vertical vibration from 10 to 60 Hz with varying centre of pressure conditions.

Modelling the foot-ankle system (FAS) while exposed to foot-transmitted vibration (FTV) is essential for designing inhibition methods to prevent the effects of vibration-induced white-foot. K-means analysis was conducted on a data set containing vibration transmissibility from the floor to 24 anatomical locations on the right foot of 21 participants. The K-means analysis found three locations to be sufficient for summarising the FTV response. A three segment, four degrees-of-freedom lumped parameter model of the FAS was designed to model the transmissibility response at three locations when exposed to vertical vibration from 10 to 60 Hz. Reasonable results were found at the ankle, midfoot, and toes in the natural standing position (mean-squared error (ε) = 0.471, 0.089, 0.047) and forward centre of pressure (COP) (ε = 0.539, 0.058, 0.057). However, when the COP is backward, the model does not sufficiently capture the transmissibility response at the ankle (ε = 1.09, 0.219, 0.039). Practitioner summary The vibration transmissibility response of the foot-ankle system (FAS) was modelled with varying centre of pressure (COP) locations. Modelling the FAS using three transmissibility locations and two foot segments (rearfoot and forefoot) demonstrated reasonable results in a natural standing and forward COP position to test future intervention strategies. Abbreviations: COP: centre of pressure; DOF: degrees-of-freedom; FAS: foot-ankle system; FTV: foot-transmitted vibration; HAVS: hand-arm vibration syndrome; LDV: laser Doppler vibrometer; LP: lumped-parameter; VWT: vibration-induced white-toes; WBV: whole-body vibration.

Use of Machine Learning and Wearable Sensors to Predict Energetics and Kinematics of Cutting Maneuvers.

Changes of directions and cutting maneuvers, including 180-degree turns, are common locomotor actions in team sports, implying high mechanical load. While the mechanics and neurophysiology of turns have been extensively studied in laboratory conditions, modern inertial measurement units allow us to monitor athletes directly on the field. In this study, we applied four supervised machine learning techniques (linear regression, support vector regression/machine, boosted decision trees and artificial neural networks) to predict turn direction, speed (before/after turn) and the related positive/negative mechanical work. Reference values were computed using an optical motion capture system. We collected data from 13 elite female soccer players performing a shuttle run test, wearing a six-axes inertial sensor at the pelvis level. A set of 18 features (predictors) were obtained from accelerometers, gyroscopes and barometer readings. Turn direction classification returned good results (accuracy > 98.4%) with all methods. Support vector regression and neural networks obtained the best performance in the estimation of positive/negative mechanical work (coefficient of determination R2 = 0.42-0.43, mean absolute error = 1.14-1.41 J) and running speed before/after the turns (R2 = 0.66-0.69, mean absolute error = 0.15-018 m/s). Although models can be extended to different angles, we showed that meaningful information on turn kinematics and energetics can be obtained from inertial units with a data-driven approach.

Standing centre of pressure alters the vibration transmissibility response of the foot.

Vibration-white foot as an occupational disease has underscored the need to better understand the vibration response of the foot. While vibration transmissibility data exist for a natural standing position, it is anticipated that weight distribution will affect the response. The purpose of this study was to determine the effects of changes in centre of pressure (COP) on the foot's biomechanical response. Twenty-one participants were exposed to vertical vibration of 30 mm/s, with a sine sweep from 10-200 Hz. Z-axis (vertical) vibration was measured at 24 locations on the right foot, with the COP shifted forward or toward the heel. A mixed model analysis at each location revealed significant differences (p < .001) in the transmissibility response when the COP was altered to the forefoot and rearfoot. In general, the peak frequency of the average vibration response increased for a region of the foot when the COP was shifted toward that region. Practitioner Summary: Altering the centre of pressure location resulted in changes in the transmission of vibration through the foot. The forward lean position was associated with the greatest amplitude of vibration transmissibility at the toes. This information is relevant for clinicians studying vibration-induced white-foot and engineers designing protective equipment.

Biomechanical response of the human foot when standing in a natural position while exposed to vertical vibration from 10-200 Hz.

Exposure to foot-transmitted vibration (FTV) can lead to pain and numbness in the toes and feet, increased cold sensitivity, blanching in the toes, and joint pain. Prolonged exposure can result in a clinical diagnosis of vibration-induced white foot (VIWFt). Data on the biomechanical response of the feet to FTV is limited; therefore, this study seeks to identify resonant frequencies for different anatomical locations on the human foot, while standing in a natural position. A laser Doppler vibrometer was used to measure vertical (z-axis) vibration on 21 participants at 24 anatomical locations on the right foot during exposure to a sine sweep from 10-200 Hz with a peak vertical velocity of 30 mm/s. The most notable differences in the average peak frequency occur between the toes (range: 99-147 Hz), midfoot (range: 51-84 Hz) and ankle (range: 16-39 Hz). Practitioner Summary: The biomechanical response of the human foot exposed to foot-transmitted vibration, when standing in a natural position, was measured for 21 participants. The foot does not respond uniformly; the toes, midfoot, and ankle regions need to be considered independently in future development of isolation strategies and protective measures.

Apparent mass matrix of standing subjects exposed to multi-axial whole-body vibration.

This paper describes the experimental characterisation of the apparent mass matrix of eight male subjects in standing position and the identification of nonlinearities under both mono-axial and dual-axis whole-body vibration. The nonlinear behaviour of the response was studied using the conditioned response techniques considering models of increasing complexity. Results showed that the cross-axis terms are comparable to the diagonal terms. The contribution of the nonlinear effects are minor and can be endorsed to the change of modal parameters during the tests. The nonlinearity generated by the vibration magnitude is more evident in the subject response, since magnitude-dependent effects in the population are overlaid by the scatter in the subjects' biometric data. The biodynamic response is influenced by the addition of a secondary vibration axis and, in case of dual-axis vibrations, the overall magnitude has a marginal contribution. Practitioner Summary: We have measured both the diagonal and cross-axis elements of the apparent mass matrix. The effect of nonlinearities and the simultaneous presence of vibration along two axes are smaller than the inter-subject variability.

Whole-body vibration exposure in sport: four relevant cases.

This study investigates the whole-body vibration exposure in kite surfing, alpine skiing, snowboarding and cycling. The vibration exposure was experimentally evaluated following the ISO 2631 guidelines. Results evidenced that the most critical axis is the vertical one. The weighted vibration levels are always larger than 2.5 m/s(2) and the vibration dose values are larger than 25 m/s(1.75). The exposure limit values of the EU directive are reached after 8-37 min depending on the sport. The vibration magnitude is influenced by the athletes' speed, by their skill level and sometimes by the equipment. The large vibration values suggest that the practice of sport activities may be a confounding factor in the aetiology of vibration-related diseases. PRACTITIONER SUMMARY: The vibration exposure in some sports is expected to be large, but has never been quantified in the literature. Results of experiments performed in cycling, alpine and water sports outlined vibration levels exceeding the EU standard limit values.

Analysis of non-linear response of the human body to vertical whole-body vibration.

The human response to vibration is typically studied using linear estimators of the frequency response function, although different literature works evidenced the presence of non-linear effects in whole-body vibration response. This paper analyses the apparent mass of standing subjects using the conditioned response techniques in order to understand the causes of the non-linear behaviour. The conditioned apparent masses were derived considering models of increasing complexity. The multiple coherence function was used as a figure of merit for the comparison between the linear and the non-linear models. The apparent mass of eight male subjects was studied in six configurations (combinations of three vibration magnitudes and two postures). The contribution of the non-linear terms was negligible and was endorsed to the change of modal parameters during the test. Since the effect of the inter-subject variability was larger than that due to the increase in vibration magnitude, the biodynamic response should be more meaningfully modelled using a linear estimator with uncertainty rather than looking for a non-linear modelling.

Relation of digital arterial dysfunction to alternative frequency weightings of hand-transmitted vibration.

This study compared the relative performance of alternative frequency weightings of hand-transmitted vibration (HTV) to predict the extent of cold-induced vasoconstriction in the digital arteries of HTV workers. The cold response of digital arteries was related to measures of daily vibration exposure expressed in terms of r.m.s. acceleration magnitude normalised to an 8-h day, frequency weighted according to either the frequency weighting Wh defined in international standard ISO 5349-1:2001 (Ah(8) in ms-2 r.m.s.) or the hand-arm vascular frequency weighting Wp proposed in the ISO Technical Report 18570:2007 (Ap(8) in ms-2 r.m.s.). The measure of daily vibration exposure constructed with the frequency weighting Wp (Ap(8)) was a better predictor of the cold response of the digital arteries in the HTV workers than the metric derived from the conventional ISO frequency weighting Wh (Ah(8)). This finding suggests that a measure of daily vibration exposure constructed with the vascular weighting Wp, which gives more weight to intermediate- and high-frequency vibration (31.5-250 Hz), performed better for the prediction of cold induced digital arterial hyperresponsiveness than that obtained with the frequency weighting Wh recommended in ISO 5349-1 which gives more importance to lower frequency vibration (≤16 Hz).

The relationship between jump and sprint performance in preschool children.

BACKGROUND: Physical activity may help prevent the development of adverse health disorders in children. Thus, it is fundamental to assess key physical skills, such as jumping and running, from an early age. Several studies proposed test batteries to evaluate these motor skills in preschoolers, but no research studied their association. Therefore, this study aimed to evaluate the relationship between jump performance, including force production parameters, and sprint performance in preschool children. METHODS: Twenty-nine preschoolers, aged 4 to 5, underwent assessments, including countermovement jumps (CMJ) and standing long jumps (SLJ) on a force plate. Then they performed a 10-meter linear sprint assessed using photocells. RESULTS: Regression models revealed that SLJ distance emerged as a significant predictor (R2=49.3%, P<0.001) of sprint horizontal velocity, while, for sprint momentum (R2=34.3%), both SLJ distance (P=0.004) and SLJ vertical peak force (P=0.036) were found to be significant predictors. CONCLUSIONS: The findings showed that short-distance (i.e., 10 m) linear sprint performance, both velocity and momentum, in preschoolers may be predicted mainly using SLJ assessment. These findings underscore the importance of early motor skill development in shaping physical abilities and their potential relationship in preschool children.

Bioimpedance Vector Analysis-Derived Body Composition Influences Strength and Power in Alpine Skiers.

Purpose: Alpine ski racing is a complex sport where no single factor can exclusively account for performance. We assessed body composition, using bioelectrical impedance vector analysis (BIVA), and our purpose was to study its influence on the strength and power profile of young alpine skiers. Methods: Anthropometric measurements and advanced BIVA parameters were recorded on eighteen alpine ski racers (6 females/12 males; 17.0 ± 1.3 years; 172.2 ± 9.3 cm; 68.5 ± 9.8 kg). Dynamic force and power were assessed using countermovement jumps (CMJ), while maximal isometric strength was evaluated for hip flexion-extension and abduction-adduction movements. Stepwise regression models examined the relationship between BIVA-derived parameters and strength/power variables. Results: Body cellular mass (BCM) positively related to jump height (p = .021, R2 = 74%), jump momentum (p < .001, R2 = 89%), reactive strength index modified (p = .026, R2 = 75%) and peak concentric power (p < .001, R2 = 82%), while intracellular water (ICW) related to peak eccentric (p < .001, R2 = 76%) and concentric force (p < .001, R2 = 77%) as well as to concentric rate of force development together with the phase angle (PhA) (p = .008, R2 = 79% and R2 = 0.015). Regarding isometric assessment, ICW was a significant predictor for all four movement directions, and PhA contributed to hip adduction strength. Conclusions: Body composition, particularly BCM and ICW, significantly predict force- and power-related factors in young alpine skiers.

Effects of mediolateral whole-body vibration during gait with additional cognitive load.

Whole-body vibration (WBV) may increase musculoskeletal disorder risk among workers standing on vibrating surfaces for prolonged periods. Limited studies were conducted to comprehend WBV impact on individuals engaged in dynamic activities. This study explored the effects of different horizontal WBV frequencies on gait parameters, lower limb kinematics, and the cognitive response of healthy subjects. Forty participants walked at constant speed on a treadmill mounted on a horizontal shaker providing harmonic vibration with an amplitude of 1 m/s2 and frequencies 2-10 Hz, with inversely proportional amplitudes. A Psychomotor Vigilance Test measured reaction time while a motion capture system recorded walking kinematics. ANOVA results revealed no significant impact of vibration frequencies on the reaction time. At 2 Hz, alterations in gait spatiotemporal parameters were significant, with reduced stride length, stride time, step length, and stance time and increased step width and cadence. Similarly, gait variability measured by standard deviation and coefficient of variation significantly increased at 2 Hz compared to the other conditions. Comparably, kinematic time series analyzed through statistical parametric mapping showed significant adjustments in different portions of the gait cycle at 2 Hz, including increased hip abduction and flexion, greater knee flexion around the heel strike, and augmented ankle dorsiflexion. Participants exhibited gait kinematic variations, mainly at 2 Hz, where the associated mediolateral displacement was higher, as a plausible strategy to maintain stability and postural control during perturbed locomotion. These findings highlight individuals' complex biomechanical adaptations in response to horizontal WBV, especially at lower frequencies, under dual-task conditions.

Advancements in Sensor Technologies and Control Strategies for Lower-Limb Rehabilitation Exoskeletons: A Comprehensive Review.

Lower-limb rehabilitation exoskeletons offer a transformative approach to enhancing recovery in patients with movement disorders affecting the lower extremities. This comprehensive systematic review delves into the literature on sensor technologies and the control strategies integrated into these exoskeletons, evaluating their capacity to address user needs and scrutinizing their structural designs regarding sensor distribution as well as control algorithms. The review examines various sensing modalities, including electromyography (EMG), force, displacement, and other innovative sensor types, employed in these devices to facilitate accurate and responsive motion control. Furthermore, the review explores the strengths and limitations of a diverse array of lower-limb rehabilitation-exoskeleton designs, highlighting areas of improvement and potential avenues for further development. In addition, the review investigates the latest control algorithms and analysis methods that have been utilized in conjunction with these sensor systems to optimize exoskeleton performance and ensure safe and effective user interactions. By building a deeper understanding of the diverse sensor technologies and monitoring systems, this review aims to contribute to the ongoing advancement of lower-limb rehabilitation exoskeletons, ultimately improving the quality of life for patients with mobility impairments.

Apparent mass distribution at the feet of standing subjects exposed to whole-body vibration.

This study was carried out to investigate the influence of the body posture and of the foot support on the apparent mass distribution at the feet of standing subjects exposed to whole-body vibration. The apparent mass was measured at the driving point through a capacitive pressure sensor matrix, which allowed to separate the contributions of the different foot regions. The overall value was also determined using a conventional measurement system based on piezoelectric load cells. Ten male subjects performed 15 tests with three kinds of feet supports (flat rigid, anatomic rigid and flat soft) in five different postures. Static components of the pressure measurements were exploited to identify which fraction of the weight is supported by the rearfoot, the midfoot and the forefoot in the various test configurations. Factorial design of experiments on different response variables showed that the apparent mass is affected by the posture but not by the type of feet contact surface; conversely, the presence of insoles varies with the apparent mass distribution on the different feet parts. PRACTITIONER SUMMARY: The response of standing subjects to whole-body vibration has always been considered as a global parameter measured at the driving point, neglecting the local phenomena occurring in different foot parts. We have experimentally identified the apparent mass distribution of subjects in different standing postures and with different foot supports.

Corrigendum to "Energy analysis of gait in patients with down syndrome" [Heliyon 8 (11) (November 2022) Article e11702].

[This corrects the article DOI: 10.1016/j.heliyon.2022.e11702.].

The smart body concept as a demonstration of the overarching utility and benefits of 3D avatars in retail, health and wellbeing: an accuracy study of body measures from 3D reconstruction.

Recent developments in 3D graphic technologies enable the affordable and precise reconstruction of body scanned models that can be applied in a variety of verticals, such as fashion, fitness and wellness, and healthcare. The accuracy of body measurements is a crucial element for the successful application of avatars in the following use cases: Avatars that go beyond visual representation and offer intrinsic and precise anthropometric data defined as a smart body are discussed in this paper. In particular, this paper presents the Gate technology, an innovative, autonomous, sustainable body scanner, coupled with an automatic production pipeline and the concept of avatars as smart bodies. We present an accuracy study of scanning technology for scanning inanimate objects, as well as body parts versus the ground, by using an established accuracy scanning system. The results appear to be promising and confirm the hypothesis of applying the technology to the use cases discussed as well as broadening the research to other studies and future applications.