A5074796495- Land Use and Ecosystem Services
- Remote Sensing and Land Use
- Wildlife-Road Interactions and Conservation
- Prosthetics and Rehabilitation Robotics
- Muscle activation and electromyography studies
- Stroke Rehabilitation and Recovery
- Visual Attention and Saliency Detection
- Olfactory and Sensory Function Studies
- Environmental Changes in China
- AI in cancer detection
- Remote-Sensing Image Classification
- Environmental Quality and Pollution
- Urban Heat Island Mitigation
- Wildlife Ecology and Conservation
- Urban Transport and Accessibility
- Physical Activity and Health
- Economic and Environmental Valuation
- Context-Aware Activity Recognition Systems
- Non-Invasive Vital Sign Monitoring
- COVID-19 diagnosis using AI
- Soil erosion and sediment transport
- Video Surveillance and Tracking Methods
- Impact of Light on Environment and Health
- Advanced Computing and Algorithms
- Environmental Conservation and Management
Chongqing Jiaotong University
2019-2022
Soochow University
2017-2021
Defining a reasonable and feasible watershed ecological compensation threshold is the key to protecting functions maintaining sustainable utilization of ecosystems. However, many studies have obtained values that are too high promote implementation policies. This study chose upper reaches Yangtze River as research area, taking water resources closer people’s daily needs evaluation object. Based on InVEST (Integrated Valuation Ecosystem Services Trade-offs) model grey footprint method, for...
Coronavirus disease (COVID-19) is highly contagious and pathogenic. Currently, the diagnosis of COVID-19 based on nucleic acid testing, but it has false negatives hysteresis. The use lung CT scans can help screen effectively monitor diagnosed cases. application computer-aided technology reduce burden doctors, which conducive to rapid large-scale diagnostic screening. In this paper, we proposed an automatic detection method for spatiotemporal information fusion. Using segmentation network in...
Pedestrian detection and tracking has become an important field in the of computer vision research. However, existing pedestrian algorithms have some problems, such as low accuracy poor stability due to similar background overlapped occlusion interference. Therefore, occluded method based on binocular is proposed this paper. We simulate recognition human brain use deep learning network MobileNet detect locate initial pedestrians. Then, depth introduced visual salience prior information,...
The prediction of sensor data can help the exoskeleton control system to get human motion intention and target position in advance, so as reduce human-machine interaction force. In this paper, an improved method for algorithm is proposed. Through simulation test two-link experiment, improves accuracy by 14.23 ± 0.5%, smooth. Input predicted signal into model, use calculated torque verify smoothness. results showed that predict joint angle body be used follow-up swinging legs exoskeleton.
Abstract The optimization of the landscape ecological security pattern aims to construct a suitable environment and promote harmonious development between humans nature. model for main urban area Chongqing was constructed with granularity inverse method, minimum cumulative resistance model, spatial network analysis method. We used nodes optimize by organically combining quantity structure. results were as follows: (1) optimal selecting source in study 500 m. There 220 sources total 188691.03...
The effectiveness of the measurement method human exercise energy consumption based on accelerometer has been confirmed by research institutes in recent years. Aiming at shortcomings accuracy existing algorithms for calorie consumption, this article aims importance detection stoop activity expenditure (SAEE), a new system SAEE AHRS transducer was presented. With collecting and analyzing data, calculating motion parameters algorithm developed. experimental results prove that can accurately...
This paper presents a disturbing recovery approach for lower extremity exoskeletons with passive ankle joints, which particularly works when an external disturbance is applied on the standing exoskeleton. Firstly, by analyzing mechanical structure of during its stance phase, mathematical model established using state-space equations. Then, basis LQR (Linear Quadratic Regulator) theory, novel neural network controller designed to obtain optimal solution balance control proposed. Finally,...
Inspired by the visual properties of human eyes, depth information attention is integrated into saliency detection to effectively solve problems such as low accuracy and poor stability under similar or complex background interference. Firstly, improved SLIC algorithm was used segment cluster RGBD image. Secondly, image region obtained according anisotropic center-surround difference method. Then, global feature RGB calculated colour perception rule vision. The multichannel maps were weighted...