Nowadays, with the maturity and development of Internet, cloud computing and Internet of things technology, and the wide popularization of medical / health information technology, medical / health related data is growing at an amazing speed. At the same time, the popularization and application of genomic technology and the rapid development of wearable mobile medical and mobile health technology, promote the field of health care to enter the era of big data. Traditional sports risk assessment methods can assess the risks in sports, but there are problems such as time-consuming and poor accuracy in the assessment process, which can not be used in large-scale sports assessment. A method of motion risk assessment based on big data analysis is proposed. Based on the analysis of risk factors, this paper constructs a risk assessment model of large-scale sports, and introduces multi-level superposition operation and multi-factor mediation variance method, processes the sports risk data, and realizes the sports risk assessment using BP neural network based on the big data analysis method. The experimental results show that the proposed method based on big data analysis, compared with the traditional risk assessment method, can carry out high-efficiency and accurate motion risk assessment, and can be applied to large-scale risk assessment. This system has theoretical and practical significance for guiding the theory and practice of big data driven health / disease management, promoting the transformation and upgrading of sports health management and the development of health care big data industrialization.INDEX TERMS Data analysis, big data, cloud platform, motion information, EEMD-HHT.
New impulse detection and filtering algorithm is proposed in color images. Based on fast peer group filter, the proposed filtering algorithm uses different iteration times to complete filter according to different impulse noise density. The extensive experimental results show that the proposed scheme provides better performance than many of the existing vector filters. Meanwhile, the proposed approach is simple and practical for real-time application.
In this paper, we study long-range dependence of hydrological records with high frequent and massive data set. For detecting breakpoints, we apply the Evolutionary Wavelet Spectrum (EWS) to provide a segmentation of the original time series. And rescaled range analysis (R/S) for estimating the Hurst exponent that describe the long-range dependence phenomenon are used. The results affirm that the hydrological records have long-range dependent (LRD) behaviors.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.