Search Results (8)

This study proposes a behavioral monitoring framework for newborn piglets based on SPMF-YOLO object detection and ByteTrack multi-object tracking, which enables precise quantification of early postnatal activity levels and health assessment. The method enhances small-object detection performance by incorporating the SPDConv module, the MFM module, and the NWD loss function into YOLOv11. When combined with the ByteTrack algorithm, it achieves stable tracking and maintains trajectory continuity for multiple targets. An annotated dataset containing both detection and tracking labels was constructed using video data from 10 piglet pens for evaluation. Experimental results indicate that SPMF-YOLO achieved a recognition accuracy rate of 95.3% for newborn piglets. When integrated with ByteTrack, it achieves 79.1% HOTA, 92.2% MOTA, and 84.7% IDF1 in multi-object tracking tasks, thereby outperforming existing methods. Building upon this foundation, this study further quantified the cumulative movement distance of each newborn piglet within 30 min after birth and proposed a health-assessment method based on statistical thresholds. The results demonstrated an overall consistency rate of 98.2% across pens and an accuracy rate of 92.9% for identifying abnormal individuals. The results validated the effectiveness of this method for quantifying individual behavior and assessing health status in newborn piglets within complex farming environments, providing a feasible technical pathway and scientific basis for health management and early intervention in precision animal husbandry. Full article

A three-phase interior permanent magnet (IPM) machine with 18-stator-slots/12-rotor-poles and concentrated armature winding is commercially employed as a 10 kW integrated-starter-generator in a commercial hybrid electric vehicle. For comprehensive and fair evaluation, a pair of flux-switching permanent magnet (FSPM) brushless machines, namely one stator permanent magnet flux-switching (SPM-FS) machine, and one rotor permanent magnet flux-switching (RPM-FS) machine, are designed and compared under the same DC-link voltage and armature current density. Firstly, a SPM-FS machine is designed and compared with an IPM machine under the same torque requirement, and the performance indicates that they exhibit similar torque density; however, the former suffers from magnetic saturation and low utilization of permanent magnets (PMs). Thus, to eliminate significant stator iron saturation and improve the ratio of torque per PM mass, an RPM-machine is designed with the same overall volume of the IPM machine, where the PMs are moved from stator to rotor and a multi-objective optimization algorithm is applied in the machine optimization. Then, the electromagnetic performance of the three machines, considering end-effect, is compared, including air-gap flux density, torque ripple, overload capacity and flux-weakening ability. The predicted results indicate that the RPM-FS machine exhibits the best performance as a promising candidate for hybrid electric vehicles. Experimental results of both the IPM and SPM-FS machines are provided for validation. Full article

The heavy pressure to improve CO₂ emission control in industry requires the identification of key sub-sectors and the clarification of how they mitigate CO₂ emissions through various actions. Focusing on 30 Chinese provincial regions, this study quantifies the contribution of each industrial sector to regional CO₂ mitigation by combining the logarithmic mean Divisia index with attribution analysis and extract the key sectors of CO₂ mitigation for each region. Results indicate that during 2010–2019, significant emission reduction was achieved through energy intensity (74%) in Beijing, while emission reductions were attained through industrial structure changes for Anhui (50%), Henan (45%), and Chongqing (45%). The contribution to emission reduction through energy structures is not significant. The production and supply of power and heat (PSPH) is a central factor in CO₂ mitigation through all three inhibitive factors. Petroleum processing and coking (PPC) generally contributes to emission reduction through energy structures, while the smelting and pressing of ferrous metals (SPMF) through changes in industrial structures and energy intensity. PSPH and SPMF, in most regions, have not achieved the emission peak. Except in the case of coal mining and dressing (CMD), CO₂ emissions in other key sectors have almost been decoupled from industrial development. CMD effectively promotes CO₂ mitigation in Anhui, Henan, and Hunan, with larger contribution of PPC in Tianjin, Xinjiang, Heilongjiang, and that of smelting and pressing of nonferrous metals in Yunnan and Guangxi. The findings help to better identify key sectors across regions that can mitigate CO₂ emissions, while analyzing the critical emission characteristics of these sectors, which can provide references to formulating region- and sector-specific CO₂ mitigation measures for regions at different levels of development. Full article

Background: Hypermobile lateral meniscus is difficult to diagnose with imaging due to its absence of tears or anomalies. We aimed to clarify the accuracy of the preoperative diagnosis using magnetic resonance imaging (MRI). Methods: The preoperative MRI status of the posterosuperior popliteomeniscal fascicle (sPMF), anteroinferior popliteomeniscal fascicle (iPMF), and popliteal hiatus were examined retrospectively on sagittal images in the hypermobile lateral meniscus group (n = 22) and an age- and gender-matched control group (n = 44). These statuses were evaluated by a logistic regression analysis to assess their degree of diagnostic accuracy. Results: The area under the curve (AUC) of the sPMF, iPMF, popliteal hiatus, and all three criteria combined was 0.66, 0.74, 0.64, and 0.77, respectively (low, moderate, low, and moderate accuracy, respectively). The odds ratios of the most severe type 3 forms of the sPMF, iPMF, and popliteal hiatus for hypermobile lateral meniscus were significantly high (5.50, 12.20, and 5.00, respectively). Although the diagnostic accuracy was not high enough, the significantly higher odds ratio for type 3 may indicate a hypermobile lateral meniscus. Conclusion: a definitive diagnosis of hypermobile lateral meniscus is difficult with MRI findings alone; however, MRI evaluations of the iPMF, sPMF, and the widening of popliteal hiatus can be used as an adjunct to diagnosis. Full article

(1) Background: Understanding children’s motor patterns in landing is important not only for sport performance but also to prevent lower limb injury. The purpose of this study was to analyze children’s lower limb joint angles and impact force during single-leg landings (SLL) in different types of jumping sports using statistical parametric mapping (SPM). (2) Methods: Thirty children (53.33% girls, M = 10.16 years-old, standard deviation (SD) = 1.52) divided into three groups (gymnastics, volleyball and control) participated in the study. The participants were asked to do SLLs with the dominant lower limb (barefoot) on a force plate from a height of 25 cm. The vertical ground reaction force (GRF) and lower limb joint angles were assessed. SPM{F} one-way analysis of variance (ANOVA) and SPM{t} unpaired t-tests were performed during the landing and stability phases. (3) Results: A significant main effect was found in the landing phase of jumping sport practice in GRF and joint angles. During the stability phase, this effect was exhibited in ankle and knee joint angles. (4) Conclusions: Evidence was obtained of the influence of practicing a specific sport in childhood. Child volleyball players performed SLL with lower impact force and higher knee flexion than child gymnasts. Training in specific jumping sports (i.e., volleyball and gymnastics) could affect the individual capacity to adapt SLL execution. Full article

The lack of pixel-level labeling limits the practicality of deep learning-based building semantic segmentation. Weakly supervised semantic segmentation based on image-level labeling results in incomplete object regions and missing boundary information. This paper proposes a weakly supervised semantic segmentation method for building detection. The proposed method takes the image-level label as supervision information in a classification network that combines superpixel pooling and multi-scale feature fusion structures. The main advantage of the proposed strategy is its ability to improve the intactness and boundary accuracy of a detected building. Our method achieves impressive results on two 2D semantic labeling datasets, which outperform some competing weakly supervised methods and are close to the result of the fully supervised method. Full article

Designing motion representations for 3D human action recognition from skeleton sequences is an important yet challenging task. An effective representation should be robust to noise, invariant to viewpoint changes and result in a good performance with low-computational demand. Two main challenges in this task include how to efficiently represent spatio–temporal patterns of skeletal movements and how to learn their discriminative features for classification tasks. This paper presents a novel skeleton-based representation and a deep learning framework for 3D action recognition using RGB-D sensors. We propose to build an action map called SPMF (Skeleton Posture-Motion Feature), which is a compact image representation built from skeleton poses and their motions. An Adaptive Histogram Equalization (AHE) algorithm is then applied on the SPMF to enhance their local patterns and form an enhanced action map, namely Enhanced-SPMF. For learning and classification tasks, we exploit Deep Convolutional Neural Networks based on the DenseNet architecture to learn directly an end-to-end mapping between input skeleton sequences and their action labels via the Enhanced-SPMFs. The proposed method is evaluated on four challenging benchmark datasets, including both individual actions, interactions, multiview and large-scale datasets. The experimental results demonstrate that the proposed method outperforms previous state-of-the-art approaches on all benchmark tasks, whilst requiring low computational time for training and inference. Full article

Measuring of fluid properties such as dynamic viscosity and density has tremendous potential for various applications from physical to biological to chemical sensing. However, it is almost impossible to affect only one of these properties, as dynamic viscosity and density are coupled. Hence, this paper proposes kinematic viscosity as a comprehensive parameter which can be used to study the effect of fluid properties applicable to various fluids from Newtonian fluids, such as water, to non-Newtonian fluids, such as blood. This paper also proposes an ideal microplatform, namely polymeric suspended microfluidics (SPMF³), with flow plane orthogonal to the bending plane of the structure, along with tested results of various fluids covering a wide range of engineering applications. Kinematic viscosity, also called momentum diffusivity, considers changes in both fluid intermolecular forces and molecular inertia that define dynamic viscosity and fluid density, respectively. In this study a 3D suspended polymeric microfluidic system (SPMF³) was employed to detect changes in fluid parameters such as dynamic viscosity and density during fluid processes. Using this innovative design along with theoretical and experimental results, it is shown that, in fluids, the variations of fluid density and dynamic viscosity are not easily comprehensible due to their interconnectivity. Since any change in a fluid will affect both density and dynamic viscosity, measuring both of them is necessary to identify the fluid or process status. Finally, changes in fluid properties were analyzed using simulation and experiments. The experimental results with salt-DI water solution and milk with different fat concentrations as a colloidal fluid show that kinematic viscosity is a comprehensive parameter that can identify the fluids in a unique way using the proposed microplatform. Full article