Feature Extraction Algorithm of Basketball Trajectory Based on the Background Difference Method

In order to improve the teaching quality of the free throw action process in basketball courses, this paper proposes research on the teaching of free throw action by integrating motion vector field transformation. Starting from the actual situation of the player’s free throw action, it describes the problems in the teaching process of free throw action. It elaborates on the structure and principle of the motion vector field transformation algorithm in depth. In video analysis teaching of free-throw shooting action processes, the player’s motion vectors can easily be interfered with by random noise, which needs to be optimized using smoothing processing and compensation models. Combined with the teaching objectives of free throw action, the design and development of the “basketball free throw action” microteaching by integrating the motion vector field transformation through the analysis of a university sports college basketball class of 2021 students found that from the seventh week and the eighth week of the experimental group and the control group, respectively, there is a difference of 7.8% and 13%. The ball-hitting rate is significantly different from that of the control group (<0.5). A significant difference occurred (<0.05). Although both methods have improved the free throw shooting rate of the athletes, the teaching of free throw shooting action in the basketball course integrating the motion vector field transformation performs better compared to the traditional free throw shooting action teaching mode. This study provides technical support and diversified media presentation methods for the design and development of basketball micro-courses. It constructs a feasible way to develop and innovate college basketball teaching.

Section: Introductionmentioning

confidence: 99%

Teaching the process of free throw action in basketball course based on motion vector field transformation

Li,

2024

“…For the problem of a single scale of model detection and low utilization of feature information, the anchor box of the RPN network can be redesigned based on the majority of traffic signs being circular, positive open kind of ugly pin shape of the special base [7][8]. The use of the background differential method to achieve target detection, specifically the establishment of meat waiting back easy template, by detecting the difference between the current image and the background template to get the current scene of the national standard detection [9][10][11], so as to detect the fabric yarn.…”

Section: Introductionmentioning

confidence: 99%

Research on fabric yarn detection based on improved fast R-CNN algorithm

Zhang

2023

For the current commercially available fabric yarn detection, which is based on traditional machine vision methods and relies heavily on manually designed features, an improved Faster R-CNN algorithm is proposed in this paper. In this paper, based on the Faster R-CNN algorithm, the deformable convolutional Resnet-50 network is fused to improve the learning ability of woven yarn features. By designing a multi-scale model for the detection of fine features in fabric yarns, a cascade network is introduced to improve the detection accuracy and localization accuracy of woven yarns, and an optimized loss function is constructed to reduce the effect of sample imbalance. Through the experiments of the improved fast R-CNN algorithm for fabric yarn detection, we can find that the confidence level of SSD algorithm detection is 58%, and the confidence level of the original Faster R-CNN algorithm detection is 78%, while the improved Faster R-CNN can not only accurately frame the finesse problem, but also the confidence level is as high as 97%. So this paper, based on the improved Faster R-CNN algorithm, can pinpoint the problem of fabric yarn detection, improve the learning ability of detection, and can meet the current demand for fabric yarn detection in the market.

“…Some suggestions are made for implementing the team teaching method in college basketball training to provide valuable references for other college basketball training. The literature [15] pointed out that computer vision technology has been utilized in numerous industries, including military defense, security surveillance, physical examination, and intelligent transportation. Motion target detection systems rely on fast and accurate video image segmentation to identify, localize, and analyze targets, and the trajectory and feature extraction of basketball players' shooting motion is important to improve the accuracy of basketball players' shots.…”

Section: Introductionmentioning

confidence: 99%

Application of AR technology in the study of controlled structure in basketball training

Liu

2023

To calculate the standard deviation of each attribute in the training control data, the normalized results of the standard deviation coefficient vector of basketball sports metrics are used as weights. Lagrange’s equation is used to analyze the behavioral trajectories of both players to enhance the offensive and defensive transitions of individual players. The weights used to calculate the standard deviation of each attribute of the training control data are based on the normalized results of the standard deviation coefficient vector of basketball sports metrics. It was found that the compression efficiency in the basketball sports teaching video with AR technology was 98.65%, the dominance ratio of offense-defense transition was 1.089, and the variance estimation of the disturbance term was 0.4621. Scientific training with AR technology can adjust training ideas, theories, and methods in real-time to realize the sports ideas of winning the standard and nurturing people.