A5100424456- Industrial Vision Systems and Defect Detection
- Fire Detection and Safety Systems
- Image Processing Techniques and Applications
- Infrared Target Detection Methodologies
- Digital Transformation in Industry
- Industrial Technology and Control Systems
- Advanced Measurement and Detection Methods
- Advanced Algorithms and Applications
- Advanced machining processes and optimization
- Power Systems and Technologies
- Video Surveillance and Tracking Methods
- Advanced Steganography and Watermarking Techniques
- Advanced Sensor and Control Systems
- Innovative Energy Harvesting Technologies
- Energy Harvesting in Wireless Networks
- Wireless Power Transfer Systems
- Optical measurement and interference techniques
- Robot Manipulation and Learning
- Retinal Imaging and Analysis
- Power Line Inspection Robots
- Remote Sensing and LiDAR Applications
- Advanced Chemical Sensor Technologies
- Image and Object Detection Techniques
- Forest, Soil, and Plant Ecology in China
- Hydraulic and Pneumatic Systems
China Southern Power Grid (China)
2022-2025
Suzhou Vocational Institute of Industrial Technology
2013-2025
Beijing Research Institute of Automation for Machinery Industry (China)
2023-2024
Qingdao University of Technology
2021-2024
Qingdao University of Science and Technology
2023
Qingdao Center of Resource Chemistry and New Materials
2023
Hefei University of Technology
2019-2023
Hefei University
2023
Anhui Provincial Hospital
2023
University of Science and Technology of China
2023
Digital twin is currently undergoing a significant transformation from the conceptual and theoretical research phase to implementation application phase. However, universally adaptable development platform for digital critically needed meet requirements. Specifically, publicly accessible simulation, testing, validation which can support model building, data processing, algorithm design, configuration, etc., urgently required researchers. Furthermore, developers industry, low-code that offer...
Abstract Currently, convolutional neural networks (CNNs) have made remarkable achievements in skin lesion classification because of their end-to-end feature representation abilities. However, precise is still challenging the following three issues: (1) insufficient training samples, (2) inter-class similarities and intra-class variations, (3) lack ability to focus on discriminative parts. To address these issues, we propose a deep metric attention learning CNN (DeMAL-CNN) for classification....
The vehicle routing problem with simultaneous deliveries and pickups (VRPSDP) has attracted much research interest because of the potential to provide cost savings transportation logistics operators. Several extensions VRPSDP exist. Of these extensions, split loads (SDPSLP) been proposed eliminate capacity constraints, as well allow or for a customer be into multiple visits. Although delivery pickup activities are often constrained by time windows, few studies have considered such...
In this study, a series of high-precision integrated ceramic membranes were successfully fabricated in one step by photopolymerization three-dimensional (3D) printing. These are double-layer consisting separation and support layers. As the layer, microsized patterns is made on surface, including wavy, half-cylindrical, triangular lines. The layer composed evenly distributed micron pores. By adjusting dye content slurry optical power density printer, high-resolution precision model printed....
To enhance the accuracy of CNC machine tool fault diagnosis, this study proposes an intelligent optimization method based on combination Particle Swarm Optimization (PSO) and Bacterial Foraging Algorithm (BFA), referred to as PSO-BFA. By simulating local foraging behavior bacteria, PSO-BFA algorithm demonstrates characteristics convergence, replicability, migratory properties during parameter selection, effectively improving capability fitness value model. This leads faster convergence...
The application of intelligent chip technology based on mobile terminal in the management ultra-high voltage equipment can significantly improve reliability and security operation, reduce failure rate maintenance costs. This combines advanced sensors, data processing wireless communication technologies to provide a comprehensive device condition solution through real-time monitoring, analysis remote management. Abstract:
Since noise is inevitably introduced during the measurement process of surface electromyographic (sEMG) signals, two novel methods for denoising based on variational mode decomposition (VMD) method were proposed in this work. Prior to study, there has been no literature relating how VMD applied sEMG denoising.The first uses decompose signal into multiple functions (VMFs), each which its own center frequency and narrow band, then wavelet soft thresholding (WST) VMF. This termed VMD-WST. The...
Digital watermarking is a technique used to embed an extra piece of information into multimedia signals without degrading the signal quality. For robust audio watermarking, geometrical modifications are common operations and present many challenges because they severely alter tempo or spectral structures thus cause watermark desynchronization. However, most existing algorithms lack resynchronization ability due nongeometrically-invariant nature domain. In this paper, we consider dyadic...
With the development of modern industry, manual labor has been far from meeting needs industry. In recent years,with artificial intelligence algorithms and other related field. Machine vision applied in a wide field An overview key machine system, demonstrates some applications will be proposed this paper.
ABSTRACTSupport vector machines (SVM) have received widespread attention for their excellent ability to learn, and been applied in many fields. The application of SVM weld defect defection X-ray images is receiving more attention, which has led the rapid development detection automation. In this paper, a method based on proposed. First, all potential defects are detected by novel grey-level profile analysis. Then three feature vectors extracted from used train SVM. Lastly, trained...
A rapid, effective method of identifying repeatedly frozen meat by near-infrared spectroscopy (NIRS) combined with a self-organizing competitive neural network (SCNN) model was established. total 180 samples were adopted, including hot, cold, frozen, and meats. We compared the treatment effects four pretreatment methods on spectrogram samples, namely, multiplicative scatter correction (MSC), standard normal variables (SNV), first-order differential second-order differential. The second...
For fire rescue difficulties in tunnels, and some other types of buildings, a review key technologies research into tunnel firefighting robots is given this paper. Specifically, the technical issues such as: mechanical design; drive system control systems; anti-jamming are discussed. Also, future development discussed
An effective method to calculate the statistical Mueller matrix (SMM) of suspended particles based on polarized light scattering is presented that takes advantage Stokes vectors measurement individual particles. The calculation SMM derived statistics. Experimental results Microcystis samples confirm can characterize cells different states. Then, pairwise contrast experiments indicate great prospect applied discrimination It helps find optimal incident polarization state discriminate...
We study a new consumption stimulus model that leverages mobile payment platforms to dispense massive amounts of small-value, use-it-this-week-or-lose-it digital coupons. evaluate the effects one such program in large Chinese city using novel data platform transactions 1 million participants. Exploiting participants’ rush first-come, first-served portal, we compare spending among those who won coupons lost because minor differences timing their arrival at portal. find generate an immediate...
To solve the low recognition rate of network to flame and keep accuracy, we propose an Instance segmentation model for recognizing locating flames more accurate This is improved based on deep learning Mask R-CNN, it introduces four key components:(1) After analyzing effects space channel attention, used efficient convolution attention. (2) By comparing kernel size, optimized dilated added network, (3) eliminate redundancy, reducing depth backbone while guaranteeing accuracy network. (4)...
An energy harvester is presented to scavenge ambient vibration using Terfenol-D/PMN-PT/Terfenol-D laminate magnetoelectric transducers. The uses eight magnets make up a magnetic circuit produce concentrated flux gradient, which makes the generate high power. nonlinear characteristics and electrical-output performances of at resonance are analyzed. A prototype has been fabricated tested. experimental results in agreement with analytical results. can sufficient power supply low consumption...
ABSTRACTWith the rapid development of automobile industry, demand tyre is increasing greatly. But classification implementation still not efficient and effective for industry. This paper proposes a method using marking points indicating shapes based on support vector machine (SVM). First, printed profiles are acquired by designed image acquisition device. Second, those images handled preprocessing methods denoising, segmentation, edge detection. After preprocessing, contour coordinates...
Tsaitermes ampliceps (lower termites) and Mironasutitermes shangchengensis (higher are highly eusocial insects that thrive on recalcitrant lignocellulosic diets through nutritional symbioses with gut dwelling prokaryotes eukaryotes. We used denaturing gradient gel electrophoresis a 16S rRNA clone library to investigate i) how microbial communities adapt different cellulose lignin content, ii) the differences in dominant of 2 types termites. The results indicated microbiota composition T. was...