Research on Athlete Behavior Recognition Technology in Sports Teaching Video Based on Deep Neural Network

In recent years, due to the simple design idea and good recognition effect, deep learning method has attracted more and more researchers’ attention in computer vision tasks. Aiming at the problem of athlete behavior recognition in mass sports teaching video, this paper takes depth video as the research object and cuts the frame sequence as the input of depth neural network model, inspired by the successful application of depth neural network based on two-dimensional convolution in image detection and recognition. A depth neural network based on three-dimensional convolution is constructed to automatically learn the temporal and spatial characteristics of athletes’ behavior. The training results on UTKinect-Action3D and MSR-Action3D public datasets show that the algorithm can correctly detect athletes’ behaviors and actions and show stronger recognition ability to the algorithm compared with the images without clipping frames, which effectively improves the recognition effect of physical education teaching videos.

Abnormal Driving Behavior Recognition Method Based on Smart Phone Sensor and CNN-LSTM

Accurate identification of abnormal driving behavior is very important to improve driver safety. Aiming at the problem that threshold or traditional machine learning methods are mostly used in existing studies, it is difficult to accurately identify abnormal driving behavior of vehicles, a method of abnormal driving behavior recognition based on smartphone sensor data and convolutional neural network (CNN) combined with long and short-term memory (LSTM) was proposed. Smartphone sensors are used to collect vehicle driving data, and data sets of various driving behaviors are constructed by preprocessing the data. A recognition model based on a convolutional neural network combined with a long short-term memory network was constructed to extract depth features from data sets and recognize abnormal driving behaviors. The test results show that the accuracy of the model based on CNN-LSTM can reach 95.22%, and the performance indexes can reach more than 94%. Compared with the recognition model constructed only by CNN or LSTM, this model has higher recognition accuracy.

Analysis of Digital Long Jump Take-off Wearable Sensor Monitoring System

The wearable sensor monitoring system builds a long jump take-off recognition network model based on different digital feature extraction methods (one-dimensional digital feature extraction method, two-dimensional digital feature extraction method, and feature extraction method combining one-dimensional digitization and recursion). Experimental verification and analysis are performed on the processed sample data, and the identification effects, advantages, and disadvantages of the four methods are obtained. First, the sensor behavior movement collection software is designed based on the Android system, and the collection time and frequency are specified at the same time. In addition, for the problem of multisensor behavior recognition, an effective result fusion method is proposed. In a multisensor behavior recognition system, constructing a parallel processing architecture is conducive to improving the rate of behavior recognition. To maintain or increase the rate of behavior recognition, the result fusion method plays a vital role. Finally, this paper analyzes the process of multitask behavior recognition and constructs a residual model that can effectively integrate multitask results and fully mine data information. The experimental results show that, for the monitoring of exercise volume, we use step count statistics to extract feature values that can distinguish activity types based on human motion characteristics. This paper proposes a sample autonomous learning method to find the optimal sample training set and avoid occurrence of overfitting problems. In the recognition of 11 types of long jump take-offs, the average accuracy rate reached 98.7%. The average replacement method is used to count the number of steps, which provides a data reference for the user’s daily exercise volume.

Multi-modal Data Fusion Method for Human Behavior Recognition Based on Two IA-Net and CHMM

The multi-modal data fusion method based on IA-net and CHMM technical proposed is designed to solve the problem that the incompleteness of target behavior information in complex family environment leads to the low accuracy of human behavior recognition.The two improved neural networks(STA-ResNet50、STA-GoogleNet)are combined with LSTM to form two IA-Nets respectively to extract RGB and skeleton modal behavior features in video. The two modal feature sequences are input CHMM to construct the probability fusion model of multi-modal behavior recognition.The experimental results show that the human behavior recognition model proposed in this paper has higher accuracy than the previous fusion methods on HMDB51 and UCF101 datasets. New contributions: attention mechanism is introduced to improve the efficiency of video target feature extraction and utilization. A skeleton based feature extraction framework is proposed, which can be used for human behavior recognition in complex environment. In the field of human behavior recognition, probability theory and neural network are cleverly combined and applied, which provides a new method for multi-modal information fusion.

Risky Driving Behavior Recognition Based on Vehicle Trajectory

This paper proposes a measurement of risk (MOR) method to recognize risky driving behavior based on the trajectory data extracted from surveillance videos. Three types of risky driving behavior are studied in this paper, i.e., speed-unstable driving, serpentine driving, and risky car-following driving. The risky driving behavior recognition model contains an MOR-based risk evaluation model and an MOR threshold selection method. An MOR-based risk evaluation model is established for three types of risky driving behavior based on driving features to quantify collision risk. Then, we propose two methods, i.e., the distribution-based method and the boxplot-based method, to determine the threshold value of the MOR to recognize risky driving behavior. Finally, the trajectory data extracted from UAV videos are used to validate the proposed model. The impact of vehicle types is also taken into consideration in the model. The results show that there are significant differences between threshold values for cars and heavy trucks when performing speed-unstable driving and risky car-following driving. In addition, the difference between the proportion of recognized risky driving behavior in the testing dataset compared with that in the training dataset is limited to less than 3.5%. The recognition accuracy of risky driving behavior with the boxplot- and distribution-based methods are, respectively, 91% and 86%, indicating the validation of the proposed model. The proposed model can be widely applied to risky driving behavior recognition in video-based surveillance systems.

Human Falling Recognition Based on Movement Energy Expenditure Feature

Falls in the elderly are a common phenomenon in daily life, which causes serious injuries and even death. Human activity recognition methods with wearable sensor signals as input have been proposed to improve the accuracy and automation of daily falling recognition. In order not to affect the normal life behavior of the elderly, to make full use of the functions provided by the smartphone, to reduce the inconvenience caused by wearing sensor devices, and to reduce the cost of monitoring systems, the accelerometer and gyroscope integrated inside the smartphone are employed to collect the behavioral data of the elderly in their daily lives, and the threshold analysis method is used to study the human falling behavior recognition. Based on this, a three-level threshold detection algorithm for human fall behavior recognition is proposed by introducing human movement energy expenditure as a new feature. The algorithm integrates the changes of human movement energy expenditure, combined acceleration, and body tilt angle in the process of falling, which alleviates the problem of misjudgment caused by using only the threshold information of acceleration or (and) angle change to discriminate falls and improves the recognition accuracy. The recognition accuracy of this algorithm is verified by experiments to reach 95.42%. The APP is also devised to realize the timely detection of fall behavior and send alarms automatically.