Abstract Accurate acquisition of athletes' motion trajectories under high-speed, complex environmental conditions remains a major challenge in sports biomechanics and intelligent training analysis. This study proposes a framework for capturing Games movement trajectories using the mean shift algorithm to address limitations of conventional tracking methods in dynamic training environments. A skeletal representation consisting of 16 joints and 51 degrees of freedom is constructed to model... Accurate acquisition of athletes' motion trajectories under high-speed, complex environmental conditions remains a major challenge in sports biomechanics and intelligent training analysis. This study proposes a framework for capturing Games movement trajectories using the mean shift algorithm to address limitations of conventional tracking methods in dynamic training environments. A skeletal representation consisting of 16 joints and 51 degrees of freedom is constructed to model human motion during Games training. Quaternionbased joint representation and logarithmic mapping are employed to reduce rotational singularities and enable dimensionality reduction within a linear time-invariant system. To reduce environmental parameter dependency, the probability density function in the gradient iteration framework is integrated with kernel density estimation, while colour histograms are used as stable visual features for target localisation. Additionally, Difference of Gaussian (DoG)-based feature detection and SIFT-assisted fusion tracking are introduced to enhance robustness under illumination variation and motion blur. Experimental validation demonstrates improved detection accuracy, reduced false alarm rates, and superior clustering efficiency compared with conventional estimation-based tracking approaches. The proposed framework provides a reliable, adaptive, and computationally efficient solution for acquiring Games motion trajectories without requiring external parameter tuning Read More Read Less

Abstract With the rapid development of information technology and the widespread deployment of database systems, massive volumes of enterprise data are generated and stored every day. These datasets contain valuable information that can support strategic decision-making and business intelligence. However, migrating accounting systems to cloud environments introduces significant challenges for data integrity, confidentiality, and security. Ensuring that financial data stored in distributed cloud infrastructures remains... With the rapid development of information technology and the widespread deployment of database systems, massive volumes of enterprise data are generated and stored every day. These datasets contain valuable information that can support strategic decision-making and business intelligence. However, migrating accounting systems to cloud environments introduces significant challenges for data integrity, confidentiality, and security. Ensuring that financial data stored in distributed cloud infrastructures remains accurate, complete, and tamper free has become a critical issue for modern accounting information systems. To address this problem, this study proposes a cloud-based algorithm for verifying the integrity of accounting data that integrates rough set theory with data mining techniques. The proposed approach establishes a secure and reliable cloud storage architecture capable of performing efficient data verification through parallel processing. By incorporating rough set-based attribute reduction and feature selection mechanisms, the algorithm improves computational efficiency while maintaining high accuracy in detecting data corruption or inconsistencies. In addition, the proposed system supports both single user and multi user verification mechanisms, enabling scalable data auditing in cloud environments. Experimental evaluation demonstrates that the proposed parallel verification algorithm significantly reduces communication overhead and verification time compared with traditional single-user verification schemes. The results further show that the algorithm maintains high verification accuracy while optimising computational resource utilisation. Therefore, the proposed method provides a reliable and efficient solution for safeguarding the integrity of accounting information in cloud computing environments. Read More Read Less

Abstract In recent years, biological radar systems have been increasingly utilized in diverse applications such as safety monitoring, healthcare monitoring, environmental surveillance, and intelligent sensing systems. Accurate extraction of human vital signs from radar signals remains challenging due to motion interference, environmental noise, and multi-target conditions. In this study, a convolutional neural network (CNN)- based approach for classifying biological radar signals is proposed for multi-target... In recent years, biological radar systems have been increasingly utilized in diverse applications such as safety monitoring, healthcare monitoring, environmental surveillance, and intelligent sensing systems. Accurate extraction of human vital signs from radar signals remains challenging due to motion interference, environmental noise, and multi-target conditions. In this study, a convolutional neural network (CNN)- based approach for classifying biological radar signals is proposed for multi-target vital sign detection. Principal Component Analysis (PCA) is applied to suppress motion noise and isolate meaningful physiological signals. The GoogLeNet architecture is employed to predict the number of moving targets across different environments, achieving improved accuracy. Furthermore, a novel Differential Cross-Multiplication and State Space Method (DACM-SSM) is introduced to enhance human identity identification and signal interpretation. Experimental results demonstrate that the proposed framework improves detection efficiency, mitigates interference from respiratory harmonics, and reduces data loss caused by random variations in breathing and heartbeat signals. Compared with traditional approaches, the proposed method reduces algorithmic complexity and improves computational efficiency and signal-recognition accuracy. The results confirm that CNN-based biological radar analysis can significantly enhance multi-target vital-sign monitoring and provide reliable, real-time health-monitoring capabilities. Read More Read Less