Towards edge-collaborative, lightweight and privacy-preserving classification framework by Jinbo XIONG, Yongjie ZHOU, Renwan BI, Liang WAN, Youliang TIAN

“…Aiming at the problems of data leakage of perceptual image and computational inefficiency of privacy-preserving classification framework in edge-side computing environment, a lightweight and privacy-preserving classification framework (PPCF) was proposed to supports encryption feature extraction and classification, and achieve the goal of data transmission and computing security under the collaborative classification process of edge nodes.Firstly, a series of secure computing protocols were designed based on additive secret sharing.Furthermore, two non-collusive edge servers were used to perform secure convolution, secure batch normalization, secure activation, secure pooling and other deep neural network computing layers to realize PPCF.Theoretical and security analysis indicate that PPCF has excellent accuracy and proved to be security.Actual performance evaluation show that PPCF can achieve the same classification accuracy as plaintext environment.At the same time, compared with homomorphic encryption and multi-round iterative calculation schemes, PPCF has obvious advantages in terms of computational cost and communication overhead.…”
Get full text

MSUD-YOLO: A Novel Multiscale Small Object Detection Model for UAV Aerial Images by Xiaofeng Zhao, Hui Zhang, Wenwen Zhang, Junyi Ma, Chenxiao Li, Yao Ding, Zhili Zhang

“…In addition, the model uses lightweight convolution instead of standard convolution to reduce the model’s computation. …”
Get full text

Intelligent Detection of Tomato Ripening in Natural Environments Using YOLO-DGS by Mengyuan Zhao, Beibei Cui, Yuehao Yu, Xiaoyi Zhang, Jiaxin Xu, Fengzheng Shi, Liang Zhao

“…First, to enhance feature extraction at various levels of abstraction in the input data, this paper proposes a novel segment-wise convolution module, C2f-GB. This module performs convolution in stages on the feature map, generating more feature maps with fewer parameters and computational resources, thereby improving the model’s feature extraction capability while reducing parameter count and computational cost. …”
Get full text

YED-Net: Yoga Exercise Dynamics Monitoring with YOLOv11-ECA-Enhanced Detection and DeepSORT Tracking by Youyu Zhou, Shu Dong, Hao Sheng, Wei Ke

“…A dynamic adaptive anchor mechanism and an Efficient Channel Attention (ECA) module are introduced, while the depthwise separable convolution in the C3k2 module is optimized with a kernel size of 2. …”
Get full text

Flexi-YOLO: A lightweight method for road crack detection in complex environments. by Jiexiang Yang, Renjie Tian, Zexing Zhou, Xingyue Tan, Pingyang He

“…A Global Attention Module (GAM) is integrated to improve the model's perception of global information. The AKConv convolution operation is employed to adaptively adjust the size of convolutions, further enhancing local feature capturing. …”
Get full text

A one-stage anchor-free keypoints detection model for fast electric vehicle charging port detection and pose extraction by Feifei Hou, Qiwen Meng, Xinyu Fan, Yijun Wang

“…To address these issues, this study introduces FasterEVPoints, a state-of-the-art convolutional neural network (CNN) model integrating partial convolution (PConv) with FasterNet. …”
Get full text

Fault Diagnosis Method of Permanent Magnet Synchronous Motor Demagnetization and Eccentricity Based on Branch Current by Zhiqiang Wang, Shangru Shi, Xin Gu, Zhezhun Xu, Huimin Wang, Zhen Zhang

“…Finally, to further improve diagnostic accuracy, a cascaded convolutional neural network based on dilated convolutional layers and multi-scale convolutional layers is designed as the diagnostic model. …”
Get full text

Fault Detection in Induction Machines Using Learning Models and Fourier Spectrum Image Analysis by Kevin Barrera-Llanga, Jordi Burriel-Valencia, Angel Sapena-Bano, Javier Martinez-Roman

“…This analysis introduces a new approach by demonstrating how different convolutional blocks capture particular features: the first convolutional block captures signal shape, while the second identifies background features. …”
Get full text

Distributionally Robust Day-Ahead Dispatch Optimization for Active Distribution Networks Based on Improved Conditional Generative Adversarial Network by WEI Wei, WANG Yudong, JIN Xiaolong

“…[Results] The findings demonstrate that although the day-ahead dispatch plan cost of the proposed method increases by 1.87% and 0.21% compared with the deterministic optimization （DO） and stochastic optimization （SO） methods， the integrated operation cost decreases by 5.38% and 0.46% under the worst-case scenario， respectively. …”
Get full text

LMD_YOLO: A Lightweight and Efficient Model for Pavement Defects Detection by Shuai He, Ye Yuan, Bingyang Yin

“…The model incorporates several innovations: the Diverse Branch Block enhances the detection head, improving accuracy; the mg_conv module replaces conventional convolution layers in the neck network, optimizing feature fusion without increasing computational cost; and improvements to the backbone network further enhance efficiency. …”
Get full text

An integrated approach for advanced vehicle classification. by Rui Liu, Shiyuan Wen, Yufei Xing

“…The DWAN introduces a four-level discrete wavelet transform in the convolutional neural network architecture and combines it with Convolutional Block Attention Module (CBAM) to efficiently capture multiscale feature information. …”
Get full text

Fisheye omnidirectional stereo depth estimation assisted with edge-awareness by Junren Sun, Hao Xue, Shibo Guo, Xunqi Zheng

“…Subsequently, a spherical alignment method is used to map image features from different cameras to a unified spherical coordinate system. A cost volume is built for different depth hypotheses, which is then regularized using a 3D convolutional network. …”
Get full text

Machine Learning-Based Weld Classification for Quality Monitoring by Rojan Ghimire, Rajiv Selvam

“…In this research project, a deep learning approach is proposed to classify weld defects using convolutional neural networks (CNNs) to automate the process. …”
Get full text

Hierarchical Semi-Supervised Representation Learning for Cyber Physical Social Intelligence by Na Song, Jing Yang, Xuemei Fu, Xiangli Yang, Ying Xie, Shiping Wang

“…In the context of Cyber Physical Social Intelligence (CPSI), efficiently training and inferring from samples with limited labels poses critical challenges due to the scarcity and high cost of label acquisition for big data. The aim is to attain high accuracy at minimal cost, thereby enhancing adaptation to the CPSI scenario. …”
Get full text

FE-P Net: An Image-Enhanced Parallel Density Estimation Network for Meat Duck Counting by Huanhuan Qin, Wensheng Teng, Mingzhou Lu, Xinwen Chen, Ye Erlan Xieermaola, Saydigul Samat, Tiantian Wang

“…Traditional object detection methods for meat duck counting suffer from high manual costs, low image quality, and varying object sizes. …”
Get full text

Microgrid Load Forecasting Based on Improved Long Short-Term Memory Network by Qiyue Huang, Yuqing Zheng, Yuxuan Xu

“…Secondly, the LSTM gets improved from three aspects: multilayer convolution channel, lookahead optimizer, and AM weight. …”
Get full text

YOLOv8-TEA: Recognition Method of Tender Shoots of Tea Based on Instance Segmentation Algorithm by Wenbo Wang, Yidan Xi, Jinan Gu, Qiuyue Yang, Zhiyao Pan, Xinzhou Zhang, Gongyue Xu, Man Zhou

“…Firstly, this algorithm is based on the single-stage instance segmentation algorithm YOLOv8-seg, replacing some C2f modules in the original feature extraction network with MVB, combining the advantages of convolutional neural networks (CNN) and Transformers, and adding a C2PSA module following spatial pyramid pooling (SPPF) to integrate convolution and attention mechanisms. …”
Get full text

Data-Efficient Bone Segmentation Using Feature Pyramid- Based SegFormer by Naohiro Masuda, Keiko Ono, Daisuke Tawara, Yusuke Matsuura, Kentaro Sakabe

“…The semantic segmentation of bone structures demands pixel-level classification accuracy to create reliable bone models for diagnosis. While Convolutional Neural Networks (CNNs) are commonly used for segmentation, they often struggle with complex shapes due to their focus on texture features and limited ability to incorporate positional information. …”
Get full text

Hierarchical Knowledge Transfer: Cross-Layer Distillation for Industrial Anomaly Detection by Junning Xu, Sanxin Jiang

“…By adding two additional convolutional layers to the teacher and student networks of HKT, HKT+ achieves enhanced detection capabilities at the cost of a relatively small increase in model parameters. …”
Get full text

An analytical examination of the performance assessment of CNN-LSTM architectures for state-of-health evaluation of lithium-ion batteries by Arun Jose, Sonam Shrivastava

“…This research specifically examines the potential of the convolutional neural network–long short-term memory algorithm to improve the precision of State of Health forecasts for the battery model. …”
Get full text