Figure 1 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
This paper presents an energy-optimized electronic performance tracking system (EPTS) device for analyzing the athletic movements of football players. We first develop a tiny battery-operated wearable device that can be attached to the backside of field players. In order to analyze the strategic performance, the proposed wearable EPTS device utiliz...
Contexts in source publication
Context 1
... an EPTS device can be attached to the body of football players during the match (or training) time, sensing a number of data related to the athletic and strategic performances. As exemplified in Figure 1, the EPTS device may have a positioning system assisted by numerous calibration sensors [3][4][5], collecting quantitative data such as total distance covered, peak/average speed, or other physiological data. These on-site measurements can be used for analyzing the physical workloads of each player, providing valuable insights to optimize the performance [6,7]. ...
Context 2
... the OTS and LPS show a high sampling rate and measurement accuracy, both systems typically require high-cost calibration infrastructures installed around the stadium [8]. On the other hand, the GNSS-based EPTS device is usable in any open field, whereas the position and speed are directly measured using satellite signals, as shown in Figure 1. Due to the intrinsic errors of the GNSS module [9,10], the MEMS-based inertia measurement unit (IMU) module is in general integrated into the GNSS-based EPTS device to provide more accurate information [11][12][13]. ...
Context 3
... that the optimal f G can be different for each football activity, and the proposed sensor control strategy shown in Table 6 successfully selects the valid option, reducing the measurement errors for the standardized trajectory test. For example, the proposed approach automatically sets f G = 2 Hz when the CNN operation detects the turning_slow activity, which consumes the minimum energy without increasing the distance errors, as shown in Figure 10. On the other hand, the straight-forward method reduces the sampling rates depending only on speed, which requires more samples for turning movements as shown in Figure 8a. ...
Similar publications
Stroke patients under the background of the new crown epidemic need to be home-based care. However, traditional nursing methods cannot take care of the patients’ lives in all aspects. Based on this, based on machine learning algorithms, our work combines regression models and SVM to build a smart wearable device system and builds a system predictio...
The aim of this paper is to propose a novel room-level localization approach to locate LoRa backscatter devices, which can be easily embedded into wearable devices or smartphones. The advantages of this system lie in its series of low-cost, low-power, low-complexity and long-range features. LoRa backscattering operates by alternatively connecting a...
This paper presents a novel low-cost integrated multiband antenna design customized for smartwatch applications and wearable devices. The design consists in using a broadband planar patch antenna with circular microstrip lines and a miniaturized feeding-point with a structure of 30 × 30 × 1.6 mm3, and it is easy to deploy inside the smartwatch and...
Currently, tour guides can be implemented through mobile technologiessuch as smartphones and wearable devices. The penetrationof these technologies into people’s daily lives has made it possibleto implement more sophisticated and personalized services, revolutionizingthe tourism industry. However, the process of developingsuch applications is compl...
This paper presents a solution for the Visually Impaired (VI) based on wearable devices. VI people need support or a guide to support them to locomote from one place to another. Using Wearables help users to achieve a great understanding of the surrounding environment. The proposed system is based on wearable smart glasses to support VI to locomote...
Citations
... Motion analysis becomes important for improving athlete performance and reducing athletes' injury risk. IMU (Inertial Measurement Unit) sensors which consist of three-axis accelerometers, three-axis gyroscopes, and three-axis magnetometers have been used to estimate and provide the attitude, position, and velocity of athletes [13][14][15][16][17][18][19][20][21][22][23]. The head or foot injuries of sports players can be monitored by analyzing G-impacts and reaction forces using the measured acceleration data from IMU sensors [13][14][15]. ...
... The different IMU sensor positions can be possible to provide various physical load estimates of athletes and analysis the motion of athletes, i.e., football players movement intensity information [16], runners' stride length and stride velocity, analysis at ground contacts [17], postural demands of professional soccer players [18], velocity measurements for team sports [19], and the analysis of foot swing at football kicks [20]. Deep learning techniques using IMU sensor information were also used to classify football activities [21][22][23]. ...
... Many studies on sensor technology have been performed in sports science [13][14][15][16][17][18][20][21][22][23][24][25][26][27][28]47], but few on GNSS/IMU integration have been performed [21,47]. In this study, we investigate the loosely-coupled integration of GNSS/IMU for a wearable EPTS with football players because most commercial wearable EPTS systems do not provide GNSS raw measurement data (e.g., pseudorange). ...
This study presents the filter design of GNSS/IMU integration for wearable EPTS (Electronic Performance and Tracking System) of football players. EPTS has been widely used in sports fields recently, and GNSS (Global Navigation Satellite System) and IMU (Inertial Measurement Unit) in wearable EPTS have been used to measure and provide players’ athletic performance data. A sensor fusion technique can be used to provide high-quality analysis data of athletic performance. For this reason, the integration filter of GNSS data and IMU data is designed in this study. The loosely-coupled strategy is considered to integrate GNSS and IMU data considering the specification of the wearable EPTS product. Quaternion is used to estimate a player’s attitude to avoid the gimbal lock singularity in this study. Experiment results validate the performance of the proposed GNSS/IMU loosely-coupled integration filter for wearable EPTS of football players.
... By implementing a lightweight single-pass convolutional neural network architecture with a fused information source, the detection accuracy and location tracking performance are improved compared to single-view camera images. In addition, feature extraction methods utilizing body-worn inertial measurement units (IMU) [13] and LIDAR sensors [14] have frequently been studied. However, the methods mentioned so far are not suitable for real-time object detection environments, such as football games, due to the limitations of the moving speed, distance measurement range, and lighting environment. ...
In this paper, we analyze the recognition error of the general AI recognition model and propose the structurally modified and object-augmented AI recognition model. This model has object detection features that distinguish specific target objects in target areas where players with similar shapes and characteristics overlapped in real-time football images. We implemented the AI recognition model by reinforcing the training dataset and augmenting the object class. In addition, it is shown that the recognition rate increased by modifying the model structure based on the recognition errors analysis of general AI recognition models. Recognition errors were decreased by applying the modules of HSV processing and differentiated classes (overlapped player groups) learning to the general AI recognition model. We experimented in order to compare the recognition error reducing performance with the general AI model and the proposed AI model by the same real-time football images. Through this study, we have confirmed that the proposed model can reduce errors. It showed that the proposed AI model structure to recognize similar objects in real-time and in various environments could be used to analyze football games.
... The recent research on human activity recognition and classification that was conducted by the researchers have been applied in many parts, for instance as presented in Table 1. The usage of the EFTS and IMU in paper [13] helped the authors to analyze the activities of the football players, such as remaining stationary, walking, jogging, running, slow turning, and fast turning. In references [14,15], the author conducted research by detecting the fall by using Channel State Information (CSI) to recognize the activity of falling [14] and used wearable sensors, such as an accelerometer and a guroscope to detect the fall [15]. ...
This paper describes the implementation of real time human activity recognition systems in public areas. The objective of the study is to develop an alarm system to identify people who do not care for their surrounding environment. In this research, the actions recognized are limited to littering activity using two methods, i.e., CNN and CNN-LSTM. The proposed system captures, classifies, and recognizes the activity by using two main components, a namely camera and mini-PC. The proposed system was implemented in two locations, i.e., Sekanak River and the mini garden near the Sekanak market. It was able to recognize the littering activity successfully. Based on the proposed model, the validation results from the prediction of the testing data in simulation show a loss value of 70% and an accuracy value of 56% for CNN of model 8 that used 500 epochs and a loss value of 10.61%, and an accuracy value of 97% for CNN-LSTM that used 100 epochs. For real experiment of CNN model 8, it is obtained 66.7% and 75% success for detecting littering activity at mini garden and Sekanak River respectively, while using CNN-LSTM in real experiment sequentially gives 94.4% and 100% success for mini garden and Sekanak river.
PURPOSE: We aimed to examine the effects of limiting the number of ball touches on the physical and technical performances of junior football players during small-sided games (SSGs), which are widely used to improve football-specific physical and technical performances.METHODS: Nineteen middle-school football players participated in the study and took a pretest for their physical and technical skills to be evaluated before the main experiment. During the SSG, to balance teams according to the players’ levels of physical fitness and skill, we selected players with the highest to lowest total scores and organized them in an ABBA order. Ten players who obtained the highest scores participated in the SSG once a week for 5 weeks under the limitation of a certain number of ball touches (one, two, three, four, or free touches). Players could only play with a set number of touches. Each SSG consisted of 4-min sets with 4-min breaks after each set on a pitch with a goal.RESULTS: As the number of possible touches increased, the total distance and average speed of the players increased, and the distance ratio covered by running (over 13 km/h), but not walking or jogging, also increased. Regarding technical factors, as the number of touches a player could make increased, the number of passes decreased, whereas the rates of dribbles and defensive tackles increased.CONCLUSIONS: As the number of ball touches increased during the SSG, the young players covered a greater distance with a higher speed, unlike professional players, and the frequency of skills mostly used, such as passing and dribbling during the SSG, showed different results.
The convolutional neural network (CNN) is analyzed in sports motion identification to study human motion recognition based on deep learning and smart wearable devices in sports. First, the convolution feature extraction algorithms are introduced, which include the one-dimensional (1D) convolution feature extraction algorithm, the two-dimensional (2D) convolution feature extraction algorithm, and the combination of the 1D convolution and recurrent neural network (RNN) algorithm. Then, the human motion behavior recognition is studied based on an RNN that includes simple RNN, long short-term memory network (LSTM), bilateral recurrent neural network (BLSTM), and gated recurrent unit (GRU). Finally, the 1D CNN + LSTM algorithm is chosen as the optimal algorithm through experimental analysis and comparison. Meanwhile, two kinds of sensors supported by smart wearable devices are chosen through the above methods, and a network model is constructed for human sports behavior recognition to process data collected through smart wearable devices. Smart wearable devices can operate under any circumstances. Due to low energy consumption and cost, they have been widely used in various sports events. The sensitivity and accuracy of the sensors in the smart wearable devices can be improved through the proposed 1D + LSTM algorithm, promoting their application in various sport events.
