With the widespread application of deep learning technology in image analysis, visual-based data processing and action recognition have become current research hotspots. To improve the accuracy and real-time evaluation of the quality of aerobics actions, a new lightweight OpenPose network is developed to extract key points of athletes’ skeletons. At the same time, the adaptive multi-scale optimization is carried out on the spatiotemporal graph convolutional network to grasp the temporal characteristics of actions. Finally, a new aerobics action quality evaluation model is designed by integrating Siamese network. The accuracy of the proposed model reached 96.8%, which was 5.3% higher than existing methods. The highest action evaluation matching degree was 89.7%, the highest action improvement rate was 21.34%, the highest action similarity evaluation was 92.17%, and the shortest evaluation time was 0.71 seconds. Its efficiency and effectiveness in the detection and evaluation process are significantly higher than other advanced models. From this, the proposed method for evaluating the quality of aerobics actions has strong application potential while ensuring efficiency and accuracy, and can provide certain technical support for aerobics competitions and actions training.
YangSLiZWangJ, et al.ST-GCN human action recognition based on new partition strategy. Comput Integr Manuf Syst2023; 29(12): 4040–4059. DOI: 10.13196/j.cims.2021.0453.
2.
SunLXuMGaoY, et al.Evaluating the application of using biological pulse sensor in aerobics. Wirel Netw2023; 29(8): 3567–3582. DOI: 10.1007/s11276-023-03428-z.
3.
LiuQ. Aerobics posture recognition based on neural network and sensors. Neural Comput Appl2022; 34(5): 3337–3348. DOI: 10.1007/s00521-020-05632-w.
4.
LvDYanS. Retracted article: application of visual tracking based on infrared target imaging in aerobics action evaluation system. Opt Quant Electron2024; 56(3): 329–331. DOI: 10.1007/s11082-023-06016-9.
5.
CaoT. Analysis of aerobic training posture using machine vision for body area networks. Wirel Netw2023; 29(4): 1611–1620. DOI: 10.1007/s11276-022-03123-5.
6.
ZuoNLiuJ. Research on tracking and decomposing method of aerobics movement based on machine learning. Int J Inf Commun Technol2023; 22(4): 362–376. DOI: 10.1504/IJICT.
7.
ZhangY. A novel non-maximum suppression strategy via frame bounding box smooth for video aerobics target location. Sens Imaging2022; 23(1): 29–32. DOI: 10.1007/s11220-022-00399-x.
8.
TianKXiaQ. Feature extraction of gymnastics images based on multi-scale feature fusion algorithm. Scalable Comput Pract Exp2024; 25(5): 3394–3407. DOI: 10.12694/scpe.v25i5.3190.
9.
HouZXueY. Sports training injury risk assessment combined with dynamic analysis algorithm. Mol Cell BioMech2024; 21(3): 484–493. DOI: 10.62617/mcb484.
10.
XuYWeiSYinJ. Optimization of human posture recognition based on multi-view skeleton data fusion. Inf Technol Control2024; 53(2): 542–553. DOI: 10.5755/j01.itc.53.2.36044.
11.
ChenHHuJZhangW, et al.Spatiotemporal consistency learning from momentum cues for human motion prediction. IEEE Trans Circuits Syst Video Technol2023; 33(9): 4577–4587. DOI: 10.1109/TCSVT.2023.3284013.
12.
EnkhbatAShihTKCheewaprakobkitP. Human action recognition and note recognition: a deep learning approach using STA-GCN. Sensors2024; 24(8): 2519–2524. DOI: 10.3390/s24082519.
13.
JanbiNFGhasebMAAlmazroiAA. ESTS‐GCN: an ensemble spatial–temporal skeleton‐based graph convolutional networks for violence detection. Int J Intell Syst2024; 2024(1): 3337–3346. DOI: 10.1155/2024/2323337.
14.
LiyuanL. Retracted article: application of light imaging detection based on data information visualization in aerobics training management competition simulation system. Opt Quant Electron2024; 56(2): 148–153. DOI: 10.1007/s11082-023-05771-z.
15.
TsaiMFHuangSH. Enhancing accuracy of human action recognition system using skeleton point correction method. Multimed Tools Appl2022; 81(5): 7439–7459. DOI: 10.1007/s11042-022-12000-4.
16.
JoBKimS. Comparative analysis of OpenPose, PoseNet, and MoveNet models for pose estimation in mobile devices. Trait. Signal2022; 39(1): 119–124. DOI: 10.18280/ts.390111.
17.
Van HoorenBPecasseNMeijerK, et al.The accuracy of markerless motion capture combined with computer vision techniques for measuring running kinematics. Scand J Med Sci Sports2023; 33(6): 966–978. DOI: 10.1111/sms.14319.
18.
MuTZhangCHuangM, et al.Partitioning leakage detection in water distribution systems: a specialized deep learning framework enhanced by spatial-temporal graph convolutional networks. ACS EST Water2024; 4(8): 3453–3463. DOI: 10.1021/acsestwater.4c00285.
19.
ErgeneciMCarterDKosmasP. sEMG onset detection via bidirectional recurrent neural networks with applications to sports science. IEEE Sensors J2022; 22(19): 18751–18761. DOI: 10.1109/JSEN.2022.3198882.
20.
ChowdhuryPNShivakumaraPRaghavendraR, et al.An episodic learning network for text detection on human bodies in sports images. IEEE Trans Circuits Syst Video Technol2021; 32(4): 2279–2289. DOI:10.1109/TCSVT.2021.3092713.
21.
D’AntonioETaborriJMiletiI, et al.Validation of a 3D markerless system for gait analysis based on OpenPose and two RGB webcams. IEEE Sens J2021; 21(15): 17064–17075. DOI: 10.1109/JSEN.2021.3081188.
22.
BiniRRSerrancoliGSantiagoPR, et al.Validity of neural networks to determine body position on the bicycle. Res Q Exerc Sport2023; 94(4): 905–912. DOI: 10.1080/02701367.2022.2070103.
23.
YiEByunHOhA. Development of aerobic exercise equipment using universal design: treadmill and arm ergometer. Healthcare, MDPI2022; 10(11): 2278–2279. DOI: 10.3390/healthcare10112278.
24.
FuD. Research on intelligent recognition technology of gymnastics posture based on KNN fusion DTW algorithm based on sensor technology. Int J Wireless Mobile Comput2023; 25(1): 58–67. DOI: 10.1504/IJWMC.2023.132423.
25.
OyamaYMurayamaTOhtaT. Relationship between the recognition error of dynamic postural control ability and the extent of exercise in middle-aged older women. Sport Sci Health2022; 18(4): 1293–1299. DOI: 10.1007/s11332-022-00900-9.
26.
ShiguangS. Retracted article: simulation of sports injury prevention and rehabilitation monitoring based on fiber optic sensors and machine learning algorithms. Opt Quant Electron2024; 56(4): 616–621. DOI: 10.1007/s11082-024-06321-x.
27.
AlsawadiMSRioM. Skeleton split strategies for spatial temporal graph convolution networks. Comp Mater Cont2021; 71(3): 4643–4658. DOI: 10.32604/cmc.2022.022783.
28.
WuEQTangZRXiongP, et al.ROpenPose: a rapider OpenPose model for astronaut operation attitude detection. IEEE Trans Ind Electron2021; 69(1): 1043–1052. DOI: 10.1109/TIE.2020.3048285.
29.
MokayedHQuanTZAlkhaledL, et al.Real-time human detection and counting system using deep learning computer vision techniques. Artif Intell Appl2023; 1(4): 221–229. DOI: 10.47852/bonviewAIA2202391.
30.
ZhangMZhouZDengM. Cascaded hierarchical CNN for 2D hand pose estimation from a single color image. Multimed Tools Appl2022; 81(18): 25745–25763. DOI: 10.1007/s11042-022-12780-9.
31.
BhosleKMusandeV. Evaluation of deep learning CNN model for recognition of devanagari digit. Artif Intell Appl2023; 1(2): 114–118. DOI: 10.47852/bonviewAIA3202441.
32.
LovanshiMTiwariV. Human skeleton pose and spatio-temporal feature-based activity recognition using ST-GCN. Multimed Tools Appl2024; 83(5): 12705–12730. DOI: 10.1007/s11042-023-16001-9.
