A5101509295- Remote-Sensing Image Classification
- Infrared Target Detection Methodologies
- Hydrology and Watershed Management Studies
- Advanced Image Fusion Techniques
- Smart Grid Energy Management
- IoT-based Smart Home Systems
- Remote Sensing and Land Use
- Erythrocyte Function and Pathophysiology
- Advanced Measurement and Metrology Techniques
- Thermography and Photoacoustic Techniques
- Ion Channels and Receptors
- Oil and Gas Production Techniques
- Water-Energy-Food Nexus Studies
- Advanced machining processes and optimization
- Maritime Navigation and Safety
- Signaling Pathways in Disease
- Phagocytosis and Immune Regulation
- Soil and Unsaturated Flow
- COVID-19 Clinical Research Studies
- Fish Ecology and Management Studies
- Visual Attention and Saliency Detection
- Remote Sensing in Agriculture
- Power Line Communications and Noise
- Water Systems and Optimization
- Human Mobility and Location-Based Analysis
Wuhan University
2016-2022
Renmin Hospital of Wuhan University
2022
Northeastern University
2021
China Electronics Technology Group Corporation
2020
State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing
2016-2019
Dazhou Academy of Agricultural Sciences
2013
Sichuan Academy of Forestry
2010
In recent years, with the rapid development of unmanned aerial vehicles (UAVs) and lightweight hyperspectral imaging (HSI) sensors, mini-UAV-borne remote sensing (HRS) systems have been developed demonstrate great value application potential. Compared to spaceborne airborne HSI systems, come relatively low manufacturing running costs thus become a new research focus in field HRS.
This paper focuses on the endmember extraction (EE) technique for analyzing hyperspectral images. We first prove that reconstruction errors (REs) and abundance anomalies (AAs) (abundances fail to satisfy constraints) are effective in extracting undetected endmembers. Then, according spatial continuity of objects differing from noise or outliers with a sparse distribution, endmembers assumed be located at some salient areas RE AA maps. A novel EE algorithm termed saliency-based detection...
Non-intrusive load monitoring (NILM) is a fast developing technique for appliances operation recognition in power system monitoring. At present, most NILM algorithms rely on the assumption that all fluctuations data stream are triggered by identified appliances. Therefore, of identifying unidentified still an open challenge. To pursue scalable solution to energy contemporary appliances, we propose voltage-current (V-I) trajectory enabled deep pairwise-supervised hashing (DPSH) method NILM....
With the development of smart grids, appliance-level data information plays a vital role in power consumption. Nowadays, appliance signatures detected by non-intrusive load monitoring (NILM) can be used for anomaly detection, demand response, and electricity management. These applications increase requirements accuracy identification NILM. And it has been proved that voltage–current (V–I) trajectory applied as an effective signature to represent electrical characteristics appliances with...
The leaked heat pipeline can be detected as temperature anomalies from the air-borne thermal image. Existing methods of anomaly detection are prone to generate a large quantity false alarms. Although supervised classification reduce positive rate, it requires years accumulated training data. In this paper, we use human visual system improve capabilities in district heating system. Leakage candidates selected saliency map created by image, then buffer analysis with GIS layer is used reject...
With the widespread application of AIS (Automatic Ship Identification System), ship trajectory data is being collected and becoming increasingly available. Consequently, a lot applications have become feasible that mine value from data. In this paper, based on massive data, we aim to classify two kinds stay behavior for recognizing different areas in port, namely berth anchorage. The traditional classification model mainly distinguishes moving staying state objects, but there little research...
Preferential flow is significant for its contribution to rapid response hydrologic inputs at the soil surface and unsaturated zone flow, which critical generation in rainfall–runoff (RR) models. In combination with diffuse source-responsive equations, a new model water infiltration that incorporates preferential proposed this paper. Its performance estimating moisture catchment scale was tested observed content data from Elder sub-basin of South Fork Eel River, located northern California,...
This paper focuses on the spectral unmixing technique for analyzing hyperspectral image (HSI). In this paper, we first prove that reconstruction errors and abundance anomalies (AAs, abundances are negative or greater than one) effective in measuring purity of pixels. Then, due to continuity objects space, endmembers assumed be located at some noticeable areas residual AA maps. A saliency-based endmember detection (SED) algorithm which aims iteratively extracting from maps is proposed, where...
The classified measurement of household load can make home users keep abreast the electricity consumption in their and guide them use reasonably, at same time strengthen demand side management (DSM).Load identification is base load.According to shortcomings existing which often steady state characteristics as signature, considering uniqueness transient load, a method was proposed recognize it turns on or off, then measure each based Minkowski distance used waveform active power reactive...
Diabetic cardiac fibrosis increases ventricular stiffness and promotes diastolic dysfunction. The transient receptor potential ankyrin 1 (TRPA1) has been reported to be associated with the occurrence development of various cardiovascular diseases. However, role sperm TRPA1 in diabetic cardiomyopathy (DCM) remains unclear. Here, we investigated cardioprotective effects deficiency on DCM streptozotocin (STZ)-induced rats neonatal rat fibroblasts (CFs) incubated high glucose (HG). results...