A5100411409- Advanced Neural Network Applications
- Simulation and Modeling Applications
- Advanced Algorithms and Applications
- Remote Sensing and Land Use
- Industrial Technology and Control Systems
- Advanced Image and Video Retrieval Techniques
- Nanoplatforms for cancer theranostics
- Mechanical stress and fatigue analysis
- Inertial Sensor and Navigation
- Advanced Sensor and Control Systems
- Remote-Sensing Image Classification
- Infrared Target Detection Methodologies
- Biomedical Text Mining and Ontologies
- Embedded Systems and FPGA Design
- Advanced Measurement and Detection Methods
- MicroRNA in disease regulation
- Advanced Computational Techniques and Applications
- Vehicle License Plate Recognition
- Nanoparticle-Based Drug Delivery
- Environmental Quality and Pollution
- Advanced Image Fusion Techniques
- Advanced ceramic materials synthesis
- Handwritten Text Recognition Techniques
- Aviation Industry Analysis and Trends
- Video Surveillance and Tracking Methods
Qingdao University of Science and Technology
2024-2025
Shandong Agricultural University
2025
Chinese Academy of Civil Aviation Science and Technology
2010-2024
State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing
2022-2024
Wuhan University
2016-2024
Guangzhou University of Chinese Medicine
2021-2024
Concordia University
2022-2024
Beijing University of Technology
2011-2024
Chengdu University
2024
Xuzhou Central Hospital
2024
Abstract SBFI26, an inhibitor of FABP5, has been shown to suppress the proliferation and metastasis tumour cells. However, underlying mechanism by which SBFI26 induces ferroptosis in breast cancer cells remains largely unknown. Three cell lines were treated with CCK‐8 assessed cytotoxicity. Transcriptome was performed on Illumina platform verified qPCR. Western blot evaluated protein levels. Malondialdehyde (MDA), total superoxide dismutase (T‐SOD), Fe, glutathione (GSH) oxidized (GSSG)...
License plate detection and recognition are still important challenging tasks in natural scenes. At present, most methods have favorable effect on license under restrictive conditions, of such plates shot good angle light conditions. However, for non-restrictive as dark, bright, rotated conditions etc. from the Chinese City Parking Dataset (CCPD), performance some will be significantly reduced. In order to improve accuracy unrestricted a robust model is proposed this paper, which mainly...
To explore the potential of whole-plant quinoa (WPQ) as a high-protein source for livestock feed, this study evaluated effects additives on fermentation quality and bacterial community high-moisture WPQ silage.High-moisture was ensiled with one following additives: untreated control (C), fibrolytic enzyme (E), molasses (M), LAB inoculant (L), combination (EL) (ML). The after 60 days ensiling were analysed. Naturally fermented exhibited acetic acid-type dominated by enterobacteria, low lactic...
Pavement cracks can result in the degradation of pavement performance. Due to lack timely inspection and reparation for cracks, with development safety service life be decreased. To curb detecting these accurately plays an important role. In this paper, automatic crack detection method is proposed. For achieving real-time inspection, YOLOV5 was selected as base model. small size accuracy most deep learning-based methods cannot reach a high degree. further improve those kind methods,...
In the context of increasing popularity Big Data paradigms and deep learning techniques, we introduce a novel large-scale hyperspectral imagery dataset, termed Orbita Hyperspectral Images Dataset-1 (OHID-1). It comprises 10 images sourced from diverse regions Zhuhai City, China, each boasting 32 spectral bands with spatial resolution meters spanning range 400–1000 nanometers. The core objective this dataset is to elevate performance image classification pose substantial challenges existing...
Cohort selection for clinical trials is a key step research. We proposed hierarchical neural network to determine whether patient satisfied criteria or not.We designed (denoted as CNN-Highway-LSTM LSTM-Highway-LSTM) the track 1 of national natural language processing (NLP) challenge (n2c2) on cohort in 2018. The composed 5 components: (1) sentence representation using convolutional (CNN) long short-term memory (LSTM) network; (2) highway adjust information flow; (3) self-attention reweight...
This paper presents a drone-based early forest fire detection system. Using multiple on-board aerial sensors, thermal images, RGB and distance between points drones can be captured determined from the air. To take advantage of data different sources for confirmation, both deep learning-based traditional computer vision algorithms are developed employed. The ground station designed to work collaboratively according complexity computational demands sub-modules in this By integrating sensor...
In the “2030 Agenda for Sustainable Development” proposed by United Nations, there are several Development Goals (SDGs) related to sustainable development of urbanization. Therefore, this paper combines remote sensing products and statistics data; uses entropy weight method construct a comprehensive assessment framework urbanization in Hainan Island based on 11 SDGs; conducts spatial temporal analysis from 2011 2020. The scores study area distributed spatially pattern that high north south...
Strontium (Sr) can promote the process of bone formation. To improve bioactivity, porous allograft scaffolds (ABS) were doped with Sr and mechanical strength bioactivity evaluated. Sr-doped ABS prepared using ion exchange method. The density distribution in investigated by inductively coupled plasma optical emission spectrometry (ICP-OES), X-ray photoelectron spectroscopy (XPS), energy-dispersive (EDS). Controlled release strontium ions was measured evaluated a compressive test. simulated...
In this paper, we propose a semi-supervised text classification approach for bug triage to avoid the deficiency of labeled reports in existing supervised approaches. This new combines naive Bayes classifier and expectation-maximization take advantage both unlabeled reports. trains with fraction Then iteratively labels numerous all We also employ weighted recommendation list boost performance by imposing weights multiple developers training classifier. Experimental results on Eclipse show...
Extracting relationships between chemicals and diseases from unstructured literature have attracted plenty of attention since the are very useful for a large number biomedical applications such as drug repositioning pharmacovigilance. A machine learning methods been proposed chemical-induced disease (CID) extraction due to some publicly available annotated corpora. Most them suffer time-consuming feature engineering except deep methods. In this paper, we propose novel document-level method,...
Abstract Background Whole brain radiotherapy (WBRT) can impair patients’ cognitive function. Hippocampal avoidance during WBRT potentially prevent this side effect. However, manually delineating the target area is time-consuming and difficult. Here, we proposed a credible approach of automatic hippocampal delineation based on convolutional neural networks. Methods Referring to hippocampus contouring atlas by RTOG 0933, delineated (MD) MRI data sets (3-dimensional T1-weighted with slice...
Difficult object detection and class imbalance in are the two main challenges faced by aerial image detection. objects include small objects, of scale variation with serious background interference. Class imbalances come from number different classes sampling positive negative samples. Due to these challenges, conventional models usually cannot effectively detect images, especially balance between network speed accuracy. In this paper, YOLOv3 structure was improved an method under visual...
Remote sensing images have the characteristics of a multi-scale, high resolution, and complex background, making accuracy small target detection too low. We propose an improved method based on "you-only-look-once" v5 network model to make it more suitable for detection. First, we design four-scale layer that improves depth Then, introduce complete intersection over union-non-maximum suppression (NMS) replace commonly used NMS as prediction bounding box, also considers overlap area, center...
Abstract This paper presents an adaptive attitude finite time tracking control algorithm for quadrotor unmanned aerial vehicle (UAV) in presence of actuator faults, input saturation and external disturbance. The dynamic model the UAV is characterized with quaternion representation. Subsequently, using nonsingular terminal sliding mode surface, fault‐tolerant finite‐time scheme proposed. Using Lyapunov stability analysis, it shown that proposed method able to achieve practical convergence...
Abstract This study aims to investigate the hybrid effects of carbon nanotubes (CNTs) and steel fiber (SF) on dynamic mechanical behavior ultra-high performance concrete (UHPC) under high-speed loading effects. The functionalized CNTs content varied from 0.0% 0.20%, SF volume was added 0.5% 2.0%. UHPC specimens were tested at strain rates ranging about 60 s-1 240 using split Hopkinson pressure bar. evaluated in terms stress–strain relationships, energy absorption capacity failure patterns....
We discuss magnetic actuation for Microvision's bi-axial scanners retinal scanning displays. Compared to the common side-magnet and moving-coil approach, we have designed, assembled tested a novel magnet configuration, with magnets above below moving coil. This design reduces sizes significantly without sacrificing performance, opens further improvement paths as well.
Underground coal fire is a global geological disaster that causes the loss of resources as well environmental pollution. Xinjiang, China, one regions suffering from serious underground fires. The accurate monitoring fires critical for management and extinguishment, many remote sensing-based approaches have been developed over large areas. Among them, multi-temporal interferometric synthetic aperture radar (MT-InSAR) techniques recently employed fires-related ground deformation monitoring....
The forest fire burned area is one of the most basic factors used to describe fires and plays a vital role in damage assessment. development NSSI-NDVI vegetation index triangular space method enables simultaneous calculation flammable non-photosynthetic (NPV), combustible photosynthetic (PV), incombustible bare soil (BS) fractional cover areas. This can be compensate for that was based on NDVI only by comparing before after fires, with omission NPV area. To this end, shape consistency...
Remote sensing technology has been widely used for marine monitoring. However, due to the limitations of sensor technologies and data sources, effective monitoring ships at night remains challenging. To address these challenges, our study developed SDGST, a high-resolution glimmer ship dataset from SDGSAT-1 satellite proposed detection identification method based on YOLOv5s model, Glimmer YOLO model. Considering characteristics images, model made several improvements original In particular,...