Big data analytics for image processing and computer vision technologies in sports health management

Area of Science:

• Sports Science
• Data Analytics
• Computer Vision

Background:

• Conventional sports data systems struggle with dynamic athlete health data accuracy and precise visualization.
• The increasing reliance on data in sports necessitates advanced solutions for health management.
• Big data analytics and computer vision offer a foundation for developing effective fitness and sports solutions.

Purpose of the Study:

• To present a Big Data Analytic assisted Computer Vision Model (BD-CVM) for improved sports healthcare data management.
• To enhance accuracy and precision in analyzing and visualizing athletes' health data.
• To address limitations in current systems for dynamic health data tracking.

Main Methods:

• Utilized a publicly available sports visualization dataset for athlete health and fitness analysis.
• Employed a machine learning-assisted computer vision dynamic algorithm for image featuring and classification.
• Categorized sports videos using temporal and geographical data for analysis.

Main Results:

• Demonstrated the potential of big data analytics for effective and low-error screening of sports event data.
• The proposed BD-CVM incorporates an error analysis module to ensure data processing accuracy from sports videos.
• Achieved effective image featuring and classification of sports videos.

Conclusions:

• The presented strategy can significantly improve accuracy and precision in sports data classification and visualization.
• The BD-CVM optimizes mean square error in sports data analysis.
• This approach offers a robust solution for managing and visualizing sports persons' healthcare data.