A5100786009- Fluorine in Organic Chemistry
- Advanced Measurement and Detection Methods
- Electrical and Bioimpedance Tomography
- Non-Destructive Testing Techniques
- Synthesis and Reactions of Organic Compounds
- Flow Measurement and Analysis
- Industrial Vision Systems and Defect Detection
- Target Tracking and Data Fusion in Sensor Networks
- Infrared Target Detection Methodologies
- Asymmetric Synthesis and Catalysis
- Synthesis and Biological Evaluation
- Optical measurement and interference techniques
- Advanced Image and Video Retrieval Techniques
- Advanced Measurement and Metrology Techniques
- Inertial Sensor and Navigation
- Synthesis and Catalytic Reactions
- Robotics and Sensor-Based Localization
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Optical Systems and Laser Technology
- Magnetic Properties and Applications
- Power Quality and Harmonics
- Advanced SAR Imaging Techniques
- Power Transformer Diagnostics and Insulation
- Image Retrieval and Classification Techniques
- CCD and CMOS Imaging Sensors
Tianjin University
2012-2025
Collaborative Innovation Center of Chemical Science and Engineering Tianjin
2025
Taiyuan Institute of Technology
2020-2024
China Academy of Space Technology
2020-2024
Liaoning University
2024
South China University of Technology
2021-2023
Academy of Military Medical Sciences
2022
Tianjin Municipal Engineering Design and Research Institute
2022
Donghua University
2020
Nanjing University of Finance and Economics
2020
In this paper, a sequence-to-sequence deep learning architecture based on the bidirectional gated recurrent unit (Bi-GRU) for type recognition and time location of combined power quality disturbance is proposed. Especially, proposed methodology can determine each element in input sequence, which different from existing model employing encoder–decoder network. First, sequence normalized batched. Second, features are extracted by constructing Bi-GRU neural network, where multiple layers...
Fluorination: A wide range of nitroolefins and pyrazol-5-ones undergo a sequential 1,4-addition/dearomative-fluorination transformation when treated with catalytic amount tertiary-amine—thiourea compound the terminal electrophile, N-fluorobenzenesulfonimide, to give fluorinated products in 72–95 % yield up 99:1 d.r. 98 ee. Notably, these contain adjacent tertiary α-fluoro quaternary stereocenters (see scheme).
Monitoring the multiphase flow distribution has been a challenge in industrial processes to improve control efficiency and optimize production. Electrical resistance tomography is visualization technology that can be used solve such problem. However, image reconstruction of electrical nonlinear ill-posed mathematical To this problem, supervised V-Net deep imaging method proposed. A new 33-layer network based on convolutional neural consists three sequentially connected function modules, i.e....
Electrical resistance tomography (ERT) is an effective visualization and analysis tool for multiphase flow process through array of boundary electrodes. However, the existing deep learning network image reconstruction ERT with sparse information gradient not well trained, accuracy reconstructed images cannot meet increasing demands. In order to solve this problem improve image, a new imaging algorithm that consists initial module, feature extraction residual module proposed based on dense...
In this paper, a bag of tricks is proposed to improve the precision fabric defect detection. Although general state-of-the-art convolutional neural network detection algorithm can achieve better effect, in fact, still has enough room on Therefore, we propose three further precision. Firstly, use multiscale training, which scales single input image into number images different resolutions for so as be able adapt box distribution scales. Secondly, dimension clusters method. By observing width...
Abstract With merits of good solution processability, intrinsic flexibility, etc, organic/organic interconnecting layers (ICLs) are highly desirable for tandem organic photovoltaics (OPVs). Herein, an n‐doped cross‐linked electron transport layer (ETL), named c‐NDI‐Br:PEI is developed, via a simple in situ quaternization reaction between bromopentyl‐substituted naphthalene diimide derivative (NDI‐Br) and polyethylenimine (PEI). Due to strong self‐doping, films exhibit high electrical...
Detecting keypoints on diverse objects is essential for fine-grained visual understanding and analysis. This paper introduces Enhanced Explicit Box Detection (ED-Pose++), an end-to-end framework that leverages cascade box regression to realize both conventional interactive multi-object keypoint detection. Unlike traditional one-stage methods, ED-Pose++ innovatively redefines detection as a dual-phase explicit detection, achieving unified representation optimization process. Specifically,...
Electrical resistance tomography (ERT) is an efficient technology for rapid, accurate, and real-time monitoring of the dynamic industrial process. However, due to inherent nonlinearity ill-posed, image reconstruction ERT remains a challenging problem significant importance visualization. A novel Landweber iterative network (LIRN) that combines mathematical structure method with deep learning proposed. As algorithm, proposed solves parameters selection, as method, LIRN has less dependence on...
Simple Trifluorethylierung: Die einfache Durchführung der Reaktion (siehe Schema), die unter milden Bedingungen und ohne zusätzliche Base oder Liganden abläuft, ermöglicht schnelle parallele Synthese einer Vielzahl trifluorethylierter Alkine. Experimentelle theoretische Analysen sprechen dafür, dass das Trifluormethylcarben eine konzertierte Insertion in Csp-H-Bindung des Alkins eingehen kann.
In recent years, convolutional neural networks (CNNs) have become a prominent tool for texture recognition. The key of existing CNN-based approaches is aggregating the features into robust yet discriminative description. This paper presents novel feature aggregation module called CLASS (Cross-Layer Aggregation Statistical Self-similarity) We model CNN maps across different layers, as dynamic process which carries statistical self-similarity (SSS), one well-known property texture, from input...
Abstract Artificial cells that mimic the architectural and functional characteristics of living not only shed light on physical principle life but also facilitate development in areas such as cell engineering biomedicine. Cell‐free systems carry out central dogma refers to transcription translation absence cells, offering a simplified controllable platform study cellular biological processes design principle. Many efforts have been devoted construct artificial by using cell‐free with...
Semantic segmentation of remote sensing ship targets is one the most challenging works in image processing, especially for small and multi‐scale target detection. To solve these problems, an efficient method based on convolutional neural networks (CNN) to detect proposed. This introduces attention model network enhance characteristics combines atrous convolution with traditional CNN increase receptive field. preserve information lost by pooling, proposed uses passthrough layer retain more...
It is extremely important to recognize type and locate stating-ending times of power quality disturbance for adopting corresponding measures suppress disturbances. The development machine learning artificial intelligence technology provides an effective way dealing with disturbance. In this paper, a deep method based on long short term memory sliding time window recognition location proposed. To be specific, the collected wave firstly transformed into gray scale image then model (LSTM)...
Task engagement is delined as loadings on energetic arousal (affect), task motivation, and concentration (cognition) [1]. It usually challenging expensive to label cognitive state data, traditional computational models trained with limited information for assessment do not perform well because of overlitting. In this paper, we proposed two deep (i.e., a classilier autoencoder) scarce information. We recruited 15 pilots conduct 4-h flight simulation from Seattle Chicago recorded their...
Two-phase stratified flow is ubiquitous in industrial processes, monitoring its phase interface important for improving the safety and efficiency of process. Electrical Resistance Tomography a promising non-intrusive visualization technique two-phase flow. However, some electrodes could lose their contact with liquid flow, aggravating under-determined image reconstruction solution resulting low-quality reconstructed traditional methods. In this paper, sparse batch normalization convolutional...
Virus‒host protein‒lncRNA interaction (VHPLI) predictions are critical for decoding the molecular mechanisms of viral pathogens and host immune processes. Although VHPLI interactions have been predicted in both plants animals, they not extensively studied viruses. For first time, we propose a new deep learning-based approach that consists mainly convolutional neural network bidirectional long short-term memory modules combination with transfer learning named CBIL‒VHPLI to predict viral–host...
In this paper, we introduce DINO-X, which is a unified object-centric vision model developed by IDEA Research with the best open-world object detection performance to date. DINO-X employs same Transformer-based encoder-decoder architecture as Grounding DINO 1.5 pursue an object-level representation for understanding. To make long-tailed easy, extends its input options support text prompt, visual and customized prompt. With such flexible prompt options, develop universal prompt-free...
Compass alignment is an important solution for the initial of strap-down inertial navigation system (SINS). Values parameters in compass circuit have a direct influence on performance alignment. The optimal differ from one SINS to another. Traditionally, these are mainly determined by experience and experiments that involves lot trial-and-error. It hard acquire or achieve To solve this problem, taking advantages genetic algorithm (GA) global searching, parallel computing robustness, GA based...
Traditional bridge crack detection methods are of high cost and risk. We propose a classification method based on climbing robot using image analysis with miniature camera mounted the to collect images. First, motion blur acquired is removed by Wiener filtering method. Second, wavelet transform used enhance fracture in image. Third, complete recognition, surface morphology applied extract fragments then KD-tree connect them. Finally, support vector machine classify cracks series basic visual...