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.

Keeping time and rhythm by replaying a sensory-motor engram

Imagine practicing a piece of music, or a speech, solely within the mind, without any sensory input or motor output. Our ability to implement dynamic internal representations is key for successful behavior, yet how the brain achieves this is not fully understood. Here we trained primates to perceive, and internally maintain, rhythms of different tempos and performed large-scale recordings of neuronal activity across multiple areas spanning the sensory-motor processing hierarchy. Results show that perceiving and maintaining rhythms engage multiple brain areas, including visual, parietal, premotor, prefrontal, and hippocampal regions. Each area displayed oscillatory activity that reflected the temporal and spatial characteristics of an internal metronome which flexibly encoded fast, medium, and slow tempos on a trial-by-trial basis. The presence of widespread metronome-related activity across the brain, in the absence of stimuli and overt actions, is consistent with the idea that time and rhythm are maintained by a mechanism that internally replays the stimuli and actions that define well-timed behavior.

Modelling Dynamic Hydrological Connectivity in the Zoigê Area (China) Based on Multi-Temporal Surface Water Observation

The sustainability of wetlands is threatened by the past and present land use practices. Hydrological connectivity is one of the most important aspects to consider for wetland rehabilitation planning purposes. Circuit theory and connectivity indices can be used to model and assess hydrological connectivity. The aim of this study was to assess spatiotemporal variation in the hydrological connectivity of the Zoigê area from 2000–2019 using both methods. The study area contains a Ramsar wetland of international importance, namely the Sichuan Ruoergai Wetland National Nature Reserve. We used a global surface water observation product as the major input for both methods, and then analyzed the temporal and spatial characteristics, in terms of important components and patches. We found that the overall connectivity has increased slightly in the last 20 years, while the probability of connection between patches of surface water has increased significantly. Important components and patches represent steppingstone habitat for the dispersal of organisms in the landscape. The main determinants of hydrological connectivity are mostly human oriented, predominantly a decrease in large livestock population size and population increase.

Desertification and Its Control along the Qinghai-Tibet Railway

The Qinghai-Tibet Railway is a magnificent project in the twenty first century. However, the problem of land desertification has arisen during the operation of the railway. Many sections of the railway roadbed are buried by sand. The ecological safety along the railway and the safe operation of the railway have attracted worldwide attention. This chapter will focus on the current situation of desertification along the Qinghai-Tibet Railway, such as key desertification sections and the temporal and spatial characteristics of the occurrence of desertification. At the same time, it introduces the characteristics of the dynamic conditions of railway desertification and the source of sand material. It is divided into two parts: biological measures and engineering measures to introduce desertification control along the railway. The biological measures focus on the selection of Lolium perenne, Festuca sinensi, Elymus breviaristatus, Elymus nutans and Poa crymophila, and other alpine native sand-fixing plant materials. The engineering measures will introduce the railway desertification comprehensive prevention and control technology system that combines solidification, resistance, and transportation.

Prediction of Air Pollutant Concentration Based on One-Dimensional Multi-Scale CNN-LSTM Considering Spatial-Temporal Characteristics: A Case Study of Xi’an, China

Air pollution has become a serious problem threatening human health. Effective prediction models can help reduce the adverse effects of air pollutants. Accurate predictions of air pollutant concentration can provide a scientific basis for air pollution prevention and control. However, the previous air pollution-related prediction models mainly processed air quality prediction, or the prediction of a single or two air pollutants. Meanwhile, the temporal and spatial characteristics and multiple factors of pollutants were not fully considered. Herein, we establish a deep learning model for an atmospheric pollutant memory network (LSTM) by both applying the one-dimensional multi-scale convolution kernel (ODMSCNN) and a long-short-term memory network (LSTM) on the basis of temporal and spatial characteristics. The temporal and spatial characteristics combine the respective advantages of CNN and LSTM networks. First, ODMSCNN is utilized to extract the temporal and spatial characteristics of air pollutant-related data to form a feature vector, and then the feature vector is input into the LSTM network to predict the concentration of air pollutants. The data set comes from the daily concentration data and hourly concentration data of six atmospheric pollutants (PM2.5, PM10, NO2, CO, O3, SO2) and 17 types of meteorological data in Xi’an. Daily concentration data prediction, hourly concentration data prediction, group data prediction and multi-factor prediction were used to verify the effectiveness of the model. In general, the air pollutant concentration prediction model based on ODMSCNN-LSTM shows a better prediction effect compared with multi-layer perceptron (MLP), CNN, and LSTM models.

Periodic Sea-Level Oscillation in Tokyo Bay Detected with the Tokyo-Bay Seafloor Hyper-Kilometric Submarine Deep Detector (TS-HKMSDD)

Meteorological-tsunami-like (or meteotsunami-like) periodic oscillation was muographically detected with the Tokyo-Bay Seafloor Hyper-Kilometric Submarine Deep Detector (TS-HKMSDD) deployed in the underwater highway called the Trans-Tokyo Bay Expressway or Tokyo Bay Aqua-Line (TBAL). It was detected right after the arrival of the 2021 Typhoon-16 that passed through the region 400 km south of the bay. The measured oscillation period and decay time were respectively 3 hours and 10 hours. These measurements were found to be consistent with previous tide gauge measurements. Meteotsunamis are known to take place in bays and lakes, and the temporal and spatial characteristics of meteotsunamis are similar to seismic tsunamis. However, their generation and propagation mechanisms are not well understood. The current result indicates that a combination of muography and trans-bay or trans-lake underwater tunnels will offer an additional tool to measure meteotsunamis at locations where tide gauges are unavailable.

Spatiotemporal variations of agricultural water footprint and its economic benefits in Xinjiang, northwestern China

Agriculture is the largest water user and is the main driving force behind water stress in Xinjiang, northwestern China. In this study, the water footprint (WF) (blue, green and gray WF) of main crop production and their temporal and spatial characteristics in Xinjiang were estimated in 2006, 2010, 2014 and 2018. The blue water footprint deficit (BWFd) was conducted and food productivity and economic benefits of WF were also analyzed via the water consumption per output value (food productivity and economic benefits). The results reveal that the WF increased from 22.75 to 44.16 billion m3 during 2006–2018 in Xinjiang, of which cotton, corn and wheat are main contributors of WF. In terms of different regions, corn has the largest WF in north Xinjiang and cotton has the largest WF in south and east Xinjiang. The BWFd broadened from − 11.51 to + 13.26 billion m3 in Xinjiang with the largest increased BWFd in Kashgar (from − 3.35 to 1.40 billion m3) and Aksu (from − 2.92 to 2.23 billion m3) of south Xinjiang and in Shihezi (from − 0.11 to 2.90 billion m3) of north Xinjiang. In addition, the water footprint food productivity does not well correspond with the water footprint economic benefits in prefectures of Xinjiang. It means we should consider the food yields priority and economic benefits priority to formulate a scientific and effective supervisor mode to realize the sustainable management of agricultural water in prefectures of Xinjiang.

Study on the traveling magnetic field water purifier

The article discusses a new technology for Uzbekistan of purifying drinking water from mechanical particles. This is achieved due to the generated traveling magnetic field of the electric winding wound on the metal cylinder. An analysis of the resulting mechanical attraction forces in the space of a cylindrical purifier is given. Mathematical expressions are given to calculate these forces. The analysis of the obtained graphs of temporal and spatial characteristics is presented. The values of currents allow effectively implementing the process of water purification from mechanical impurities is determined.

Analysis on Operation Risk of Continuous Sunken Tunnel on Expressway

Based on the number of 519 tunnel risk incidents that occurred in the two tunnels of the Qifu Tunnel and the Zhongcun Tunnel during the period from 2018 to 2019 of Guangming Expressway, this paper studies the temporal and spatial characteristics and the risk distribution of the sunken continuous tunnel during the operation period of the expressway. Type of risk event. The results show that January, May, June and September of the year, as well as 14:00–16:00 and 16:00–18:00 during the day are periods of high tunnel risk; at the entrance section of continuous tunnels, Compared with other locations, the number of risk events in the transition section and the open section with sudden environmental changes and gradient changes is more; the types of risk events include safety hazards, roadblocks, vehicle failures, rear-end collisions, and equipment failures. The main types are vehicle failures. There are certain differences in the east-west direction. There are more vehicle breakdowns in the east-bound direction, more roadblocks in the west-bound direction, and more rear-end collisions in the east-bound direction. The main types of risk events are cars and trucks. Both cars and trucks have major risk event types. It is a vehicle failure. In rear-end collisions, small cars account for 65% of the risk models; risk identification methods include gun patrol discovery, road administration reporting, etc., of which gun patrol discovery is the most important identification method, accounting for 65% of the total. Through the analysis of the risk event characteristics of the sunken continuous tunnel of the expressway, it provides reference opinions for perfecting the research deficiencies in related fields in our country.