A5101898207- Video Analysis and Summarization
- Human Pose and Action Recognition
- Radiomics and Machine Learning in Medical Imaging
- Advanced Image and Video Retrieval Techniques
- CCD and CMOS Imaging Sensors
- Advanced Neural Network Applications
- Image Retrieval and Classification Techniques
- Robotics and Sensor-Based Localization
- Multimodal Machine Learning Applications
- AI in cancer detection
- Topic Modeling
- Machine Learning and ELM
- Domain Adaptation and Few-Shot Learning
- Robot Manipulation and Learning
- Geoscience and Mining Technology
- Generative Adversarial Networks and Image Synthesis
- Energy Load and Power Forecasting
- MRI in cancer diagnosis
- Advanced Vision and Imaging
- Advanced Decision-Making Techniques
- Tensor decomposition and applications
- Image Processing and 3D Reconstruction
- Music and Audio Processing
- Image Processing Techniques and Applications
- Advanced Algorithms and Applications
Shanghai University
2022-2024
China University of Geosciences
2024
Hunan Xiangdian Test Research Institute (China)
2023
State Grid Hunan Electric Power Company Limited
2023
University of Iowa
2021-2023
Qiqihar Medical University
2021-2023
Weifang University
2023
Yunnan University
2023
University of Bristol
2018-2022
Bristol Robotics Laboratory
2018-2022
Aspect-based sentiment analysis (ABSA) aims to detect the targets (which are composed by continuous words), aspects and polarities in text. Published datasets from SemEval-2015 SemEval-2016 reveal that a polarity depends on both target aspect. However, most of existing methods consider predicting either or but not both, thus they easily make wrong predictions polarities. In particular, where is implicit, i.e., it does appear given text, do work. To tackle these limitations ABSA, this paper...
This paper presents a novel pothole detection approach based on single-modal semantic segmentation. It first extracts visual features from input images using convolutional neural network. A channel attention module then reweighs the to enhance consistency of different feature maps. Subsequently, we employ an atrous spatial pyramid pooling (comprising convolutions in series, with progressive rates dilation) integrate context information. helps better distinguish between potholes and undamaged...
Abstract Few studies have addressed radiomics based differentiation of Glioblastoma (GBM) and intracranial metastatic disease (IMD). However, the effect different tumor masks, comparison single versus multiparametric MRI (mp-MRI) or select combination sequences remains undefined. We cross-compared multiple machine learning (ML) models using mp-MRI to determine optimized configurations. Our retrospective study included 60 GBM IMD patients. Forty-five combinations ML feature reduction...
Prior radiomics studies have focused on two-class brain tumor classification, which limits generalizability. The performance of in differentiating the three most common malignant tumors (glioblastoma (GBM), primary central nervous system lymphoma (PCNSL), and metastatic disease) is assessed; factors affecting model usefulness a single sequence versus multiparametric MRI (MP-MRI) remain largely unaddressed. This retrospective study included 253 patients (120 (lung brain), 40 PCNSL, 93 GBM)....
<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> Interaction and integration of multimodality media types such as visual, audio, textual data in video are the essence semantic analysis. Contextual information propagation is useful for both intra- inter-shot correlations. However, traditional concatenated vector representation videos weakens power compensation among multiple modalities. In this paper, we introduce a higher-order tensor framework...
This paper presents a monocular vision based proximity estimation system using abstract features, such as corner points, blobs and edges, inputs to neural network. An experimental vehicle was built integrating the SCAMP-5 chip, micro-controller, an RC model car. The chip includes image sensor with embedded 256×256 processor SIMD array. pixel array programmed capture images run feature algorithms directly on focal plane, then digest them so that only sparse description data were read-out in...
In this study, we examine the relationship between job satisfaction and firm leverage using a sample of Chinese listed firms. We find that in "China's 100 Best Employers Award" winners during 2011–2017, is negatively associated with leverage. The effect more pronounced firms higher distress risk operating human capital intensive industries. confirm validity main findings matched series robustness checks. Overall, our results indicate can credibly demonstrate their commitment to stakeholders...
A variety of deep learning approaches have been applied to region interest (ROI) extraction, which is a fundamental task in the field remote sensing image (RSI) processing. However, unbalanced distribution positive and negative samples most RSIs greatly restricts performance these learning-based methods. In this study, data augmentation method based on variational autoencoder-multiscale generative adversarial network (VAE-MSGAN) with spatial channelwise attention (SCA) proposed balance...
China has made great efforts to fight environmental contamination along with rapid development and industrialization over the past few decades. However, effects of these nationwide measures, such as energy restructuring, on pollutant residuals in soil have not been well quantified. Polycyclic aromatic hydrocarbon (PAH) pollution is a major concern around world, PAH emissions are associated structure. Therefore, we speculated that adjustment structure may reduce content PAHs soil. To test...
Stereo vision technique has been widely used in robotic systems to acquire 3-D information. In recent years, many researchers have applied bilateral filtering stereo adaptively aggregate the matching costs. This greatly improved accuracy of estimated disparity maps. However, process whole cost volume is very time consuming and therefore resort some powerful hardware for real-time purpose. paper presents implementation fast on a state-of-the-art GPU. By highly exploiting parallel computing...
Breast cancer is a tumor disease with high incidence of female diseases, but it also that can reduce the mortality rate through early diagnosis and treatment. A method gray level co-occurrence matrix combined BP neural network proposed to improve recognition breast tumors. Using computer intelligent detect whether tumors are benign or malignant pattern problem microscopic images, which help diagnose diseases in advance, thus improving treatment effect. How accurately patients key prevention...
The identification of target identity attributes and its behavioral prediction are important means for providing command decision support in modern warfare.This paper analyzes the key steps process recognition behavior proposes a model based on Bayesian network.In simulation example, by integrating sensor information combining expert knowledge, can effectively accurately conduct battlefield situational awareness.Combined with background big data, this introduces distributed processing system...
In recent years, iron and steel industry of China has developed rapidly, surface defects recognition attracted wide attention in the field industrial inspection. Aiming at problems poor precision low speed traditional defect detection methods, we propose to use a fully learnable ensemble Extreme Learning Machines (ELMs), which is ELM-IN-ELM, for classification. The Local Binary Pattern adopted as basic feature extraction method. ELM-IN-ELM determines final classification decision by...
Human motion capturing is of great importance in video information retrieval, hence, this paper, we propose a novel approach to effectively human motions based on modified hidden markov model from multi-view image sequences. Firstly, the structure skeleton illustrated, which extended root and spine root, consists right leg, left leg spine. Secondly, our proposed system made up data training module module. In module, multi-views extracted database, feature database constructed through...