Chong Dong

An automatic two-stage method for detecting angle steel overlap anomalies using deep learning and machine vision

Jiaming Han, Ting Fang, Wensheng Liu, Zhang Wang et autres

Overlap anomaly detection is a vital task during angle steel production. However, current anomaly detection is usually carried out using visual inspection, which is inefficient and costly. Moreover, although designing machine vision solutions for non-destructive monitoring has become a trend, to the …

cn, us (code pays fourni par la source)

Three-Dimensional Attribute Modeling and Deep Mineralization Prediction of Vein 171 in Linglong Gold Field, Jiaodong Peninsula, Eastern China

Hongda Li, Zhichun Wu, Shouxu Wang, Yongfeng Wang et autres

As shallow mineral resources become increasingly depleted, the search for deep-seated orebodies has emerged as a crucial focus in modern gold exploration. This study investigates Vein 171 in the Linglong gold field, Jiaodong Peninsula, using 3D attribute modeling for deep mineralization prediction …

cn (code pays fourni par la source)

Bone marrow mesenchymal stem cells alleviate liver fibrosis after rat liver transplantation through JAK1/STAT5 pathway

Zhuyuan Si, Shengqiao Zhao, Zhixin Zhang, Tianran Chen et autres

OBJECTIVE: The effectiveness of bone marrow mesenchymal stem cells (BMSCs) in post-transplantation liver fibrosis has not been studied. The aim of this study was to investigate the effect of BMSCs on liver fibrosis and their role in the Janus-activated kinase (JAK) 1/ …

cn (code pays fourni par la source)

Bearing Fault Diagnosis Based on FFT-CNN-BiGRU-Attention

Chong Dong, Ting Fang

Aiming at the existing bearing fault diagnosis method based on one-dimensional convolutional neural network with low diagnostic accuracy in noisy environment, lack of key feature extraction and loss of timing features, this paper proposes a rolling bearing fault diagnosis method based on …

cn (code pays fourni par la source)

Motor Bearing Fault Diagnosis Based on FFT-CNN-Transformer Modeling

Chong Dong, Ting Fang

Aiming at the problems of traditional motor bearing fault diagnosis methods relying on experts' experience, difficult feature extraction and low recognition accuracy, this paper proposes an end-to-end fault diagnosis model based on FFT-CNN-Transformer. The model first utilizes the Fast Fourier Transform (FFT) …

cn (code pays fourni par la source)

A Lightweight Object Detection Algorithm for Steel Profile Type Recognition and Counting Based on YOLOv5s

Xinlong Yang, Jiaming Han, Chong Dong, Ting Fang

A lightweight object detection algorithm based on the YOLOv5s design is introduced to meet the needs of industrial sites for real-time accurate type identification, steel profile counting and fast removal of abnormally cut steel. The lightweight Ghost convolutional network is used to …

cn (code pays fourni par la source)

Motor Fault Diagnosis Based on CNN-TCN-Attention Network

Chong Dong, Ting Fang

To enhance the precision of motor fault detection further, this paper proposes a fault diagnosis model based on CNN-TCN-Attention. The model extracts the local features of the time series data through the CNN layer, captures the long-term dependence of the time series …

cn (code pays fourni par la source)

Topographic and temporal patterns of dyskinesia in multiple system atrophy with predominant parkinsonism

Hong Sun, Chong Dong, Yuan Li, Miaomiao Li et autres

A Lightweight Model for Detecting Overlapping Anomalies in Steel Sections Based on YOLOv5

Dunhui Xiao, Ting Fang, Jiaming Han, Chong Dong et autres

Aiming at the problems of overlapping anomalies of steel profiles, complex model structure and large number of model parameters in the current production process of steel profiles, an improved lightweight target detection model of yolov5s is proposed. Firstly, the MobileNetV3 network is …

cn (code pays fourni par la source)

Fault Diagnosis of Motor Bearing Based on Improved SSA Optimized VMD Combined with Convolution Neural Network

Hongkai Zhang, Ting Fang, Fanliang Meng, Jiaming Han et autres

In order to solve problems that swarm intelligence optimization algorithm is easy to fall into local optimization when optimizing the parameters of variational mode decomposition, which leads to incomplete feature extraction, which leads to low accuracy of subsequent fault diagnosis, a motor …

cn (code pays fourni par la source)

IMUSHIELD: defending against inaudible attacks via IMUs on smartphones

Zhenfei Zhang, Ping Li, Biaokai Zhu, Chong Dong et autres

Inaudible attack has brought growing concerns over security of voice assistants. With a well-designed inaudible signal, an adversary can force the voice assistant to execute commands inaudibly like “Siri, open the door”. It is challenging to defend against ultrasonic attacks without modifying …

cn (code pays fourni par la source)

TagVibra: push the limit of RFID-based vibration sensing

Jian Li, Ping Li, Biaokai Zhu, Chong Dong et autres

RFID-based mechanical vibration detection is considered a promising method for many Internet of Things (IoT) applications. However, existing methods are affected by ambient interference and leakage from the reader signature, which makes it difficult to obtain high-precision detection results. To overcome this …

cn (code pays fourni par la source)