Can Machine Learning with IMUs Be Used to Detect Different Throws and Estimate Ball Velocity in Team Handball?

Tillaar, Roland van den; Bhandurge, Shruti

doi:10.3390/s21072288

Cited by 14 publications

(3 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…14 The application of such psychological strategies is crucial for the athletes, as it promotes better coping skills during competitions. 8,9 The utilization of machine learning has gained popularity due to attention in recent years within the sporting domain, particularly in activity recognition, [15][16][17] match outcome predictions [18][19][20] and performance analysis 9,21,22 amongst others. For instance, van den Tillaar et al 23 evaluated the classification of different handball throws, apart from the prediction of ball velocity, through the employment of different supervised machine learning models.…”

Section: Introductionmentioning

confidence: 99%

“…The utilization of machine learning has gained popularity due to attention in recent years within the sporting domain, particularly in activity recognition, 15–17 match outcome predictions 18–20 and performance analysis 9,21,22 amongst others. For instance, van den Tillaar et al 23 evaluated the classification of different handball throws, apart from the prediction of ball velocity, through the employment of different supervised machine learning models. A classification accuracy (CA) between 79% and 87% was reported to be attained via the Gradient Boosting Machine, while an excellent prediction of the ball speed was obtained through a variation of the Support Vector Machine (SVM) model.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identification of high-performance volleyball players from anthropometric variables and psychological readiness: A machine-learning approach

Musa

Majeed

Suhaimi

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part P:

View full text Add to dashboard Cite

Modern indoor volleyball has evolved into a high-level strength sport and is seen as one of the most popular open-skilled team sports. The nature of the sport as an open-based skill requires players to have a high degree of both psychological skill and physical ability to cope with the sport’s externally and internally induced pace. The purposes of this study were to examine the essential basic anthropometric variables, as well as competition and practice psychological readiness, that could provide a performance edge and identify high and low-performance players based on the parameters. The anthropometric variables of height, weight, and age were assessed, while the test for performance strategies instrument was used to evaluate competition and practice psychological readiness skills of the players. The players’ performances were analyzed in real-time during a volleyball tournament. The Louvain clustering algorithm was used to determine the performance class of the players with reference to the variables evaluated. A total of 45 players were ascertained as high-performance volleyball players (HVP), while 20 players were deemed as low-performance volleyball players (LVP) via the clustering analysis technique. The logistic regression classifier was used to classify the performance of the players. Nonetheless, owing to the skewed representation between the HVP and LVP during the training of the model, the Synthetic Minority Oversampling TEchnique (SMOTE) was employed to artificially increase the minority class dataset to avoid the overfitting notion upon classification. It was shown from the study that, through the machine learning pipeline developed, an excellent identification of the HVP and LVP could be attained. The findings could be invaluable to coaches and other relevant stakeholders in team preparation and the selection of high-performance players in volleyball.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Identification of high-performance volleyball players from anthropometric variables and psychological readiness: A machine-learning approach

Musa

Majeed

Suhaimi

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part P:

View full text Add to dashboard Cite

show abstract

“…In tennis, IMU and various machine learning techniques demonstrated high accuracy in classifying shot count and type during training sessions (17). In handball, van den Tillaar et al (18) investigated the use of IMU and machine learning techniques to classify different throw types and to estimate peak ball velocity under controlled conditions. They reported that throw type could be predicted with between 80% and 87% accuracy.…”

mentioning

confidence: 99%

A Dual Inertial Measurement Unit System for Classifying Standard Overhead Drill Movements in Elite Women’s Water Polo

King,

Lewis,

Watson

et al. 2023

Medicine &Amp; Science in Sports &Amp; Exercise

View full text Add to dashboard Cite

Introduction Water polo upper limb external load monitoring cannot be currently measured accurately due to technological and methodological challenges. This is problematic as large fluctuations in overhead movement volume and intensity may affect performance and alter injury risk. Inertial measurement units (IMUs) and machine learning techniques have been shown to accurately classify overhead movements in other sports. We investigated the model accuracy and class precision, sensitivity and specificity of IMUs and machine learning techniques to classify standard overhead drill movements in elite women’s water polo. Methods Ten women’s water polo players performed standard drills of swimming, blocking, low and high intensity throwing under training conditions. Athletes wore two IMUs: one on the upper back and the other on the distal forearm. Each movement was videoed and coded to a standard overhead drill movement. IMU and coded video data were merged to verify the IMU detected activity classification of each movement to that of the video. Data were partitioned into a training and test set and used to form a decision tree algorithm. Model accuracy and class precision, sensitivity and specificity were assessed. Results IMU resultant acceleration and angular velocity values displayed drill specific values. 194 activities were identified by the model in the test set with 8 activities being incorrectly classified. Model accuracy was 95.88%. Percentage class precision, sensitivity and specificity were as follows: blocking (96.15, 86.21, 99.39), high intensity throwing (100, 100, 100), low intensity throwing (93.48, 93.48, 97.97) and swimming (94.81, 98.65, 96.67). Conclusions IMUs and machine learning techniques can accurately classify standard overhead drill movements in elite women’s water polo.

show abstract

Inverse Kinematics to Simulate Sport Movements in Virtual Environment

Sanseverino,

Genua,

Krumm

et al. 2024

Lecture Notes in Mechanical Engineering

View full text Add to dashboard Cite

Can Machine Learning with IMUs Be Used to Detect Different Throws and Estimate Ball Velocity in Team Handball?

Cited by 14 publications

References 28 publications

Identification of high-performance volleyball players from anthropometric variables and psychological readiness: A machine-learning approach

Identification of high-performance volleyball players from anthropometric variables and psychological readiness: A machine-learning approach

A Dual Inertial Measurement Unit System for Classifying Standard Overhead Drill Movements in Elite Women’s Water Polo

Inverse Kinematics to Simulate Sport Movements in Virtual Environment

Contact Info

Product

Resources

About