Results 101 to 110 of about 294 (197)

Deep‐Learned Channel Estimation for MIMO‐OFDM System by Exploiting Frequency‐Space Correlation

Electronics Letters, Volume 62, Issue 1, January/December 2026.
In massive MIMO–OFDM systems, accurate channel state information acquisition is vital for reliable transmission. Traditional methods fail to fully exploit sparse channel correlation, so an attentive residual autoencoder network is proposed. It uses an autoencoder with attention and residual connections to capture frequency–space correlation, and ...
Yiming Wei, Wenjun Jiang, Chenchen Liu, Xiaojun Yuan, Xiaojing Xu, Hua Rui   +5 more
wiley   +1 more source

Typical Industry Customers’ Demand Response Potential Evaluation Method Based on Integrated Empirical Mode Decomposition and Multi‐Head Convolutional Self‐Attention

IET Generation, Transmission &Distribution, Volume 20, Issue 1, January/December 2026.
1. An integrated empirical mode decomposition (IEMD)‐based comprehensive DR feature extraction method for typical industrial customers is proposed. This method first decomposes typical load sequences using IEMD to extract objective DR features while mitigating mode‐mixing problems and then incorporates four subjective willingness indicators to ...
Yuanqian Ma, Feng Lu, Lei Yao, Yunchu Wang, Jiaxu Geng, Zhenzhi Lin   +5 more
wiley   +1 more source

PASENet: Snowy Scene 3D Object Detection With Pillar‐Wise Attention and Semantic Enhancement

IET Image Processing, Volume 20, Issue 1, January/December 2026.
LiDAR‐based 3D object detection suffers from degraded performance in snowy conditions due to false returns and occlusions. To address the overcomes, we propose pillar‐wise attention and semantic enhancement network (PASENet), an end‐to‐end network featuring pillar‐wise attention for noise suppression and a semantic enhancement branch to improve feature
Yutian Wu, Wenwei Sun, Zuodong Zhong, Qing Li   +3 more
wiley   +1 more source

CLS‐Transformer for Home Service Robots: A Lightweight Action Recognition Model Based on Skeleton and Native Length Encoding

IET Image Processing, Volume 20, Issue 1, January/December 2026.
The paper presents a native length encoding approach alongside a human skeleton‐based action recognition model (named CLS‐Transformer) for home service robots. The model incorporates a novel cue coding strategy that captures richer feature information. The experimental results demonstrate that CLS‐Transformer achieves superior recognition accuracy and ...
Jie Shen, Yuming Zhang, Xiaoyang Li, Meng Zhou, Wenfan Wang, Dawei Zhang   +5 more
wiley   +1 more source

Pediatric Wrist Fracture Detection Using Feature Context Excitation Modules in X‐Ray Images

IET Image Processing, Volume 20, Issue 1, January/December 2026.
This study aims to propose novel neural network models serving as computer‐assisted diagnosis (CAD) tools to assist surgeons in diagnosing pediatric wrist fractures. This study integrates multiple independent blocks into the YOLOv8 model and presents three methods to enhance the model architecture, aiming to improve overall performance.
Rui‐Yang Ju, Chun‐Tse Chien, Enkaer Xieerke, Jen‐Shiun Chiang   +3 more
wiley   +1 more source

A Robust Multi‐Oriented License Plate Detector and A Derived End‐to‐End License Plate Recognizer

IET Image Processing, Volume 20, Issue 1, January/December 2026.
This is a research paper on license plate detection and recognition. A new center‐aware license plate detection and end‐to‐end license plate recognition framework is proposed for robust and efficient license plate detection and recognition under unconstrained scenarios.
Xudong Fan, Wei Zhao
wiley   +1 more source

ACT‐Agent: Affinity‐Cross Transformer for Point Cloud Registration via Reinforcement Learning

IET Image Processing, Volume 20, Issue 1, January/December 2026.
1. We proposed a point cloud registration framework combining imitation learning and reinforcement learning, which formulates the registration problem as a Markov decision process with discrete action sequences. 2. We designed the Affinity‐Cross Transformer module to enhance the point cloud feature representation and interaction capabilities. 3.
Fengguang Xiong, Haixin Gong, Qiao Ma, Yingbo Jia, Ruize Guo, Yu Cao, Ligang He, Liqun Kuang, Xie Han   +8 more
wiley   +1 more source

MLC: Enhanced Deepfake Detection Through Multi‐Level Collaborations

IET Image Processing, Volume 20, Issue 1, January/December 2026.
A novel Multi‐Level Collaborations (MLC) strategy to enhance the generalisation performance through simultaneously extracting and integrating three distinct levels of discriminative features during the encoding process. Three complementary forge clues: pixel‐level fine‐grained, region‐level facial layout, and semantic‐level deep clues, are integrated ...
Lihua Wang, Jingwei Li, Weidan Yan, Xinyu Wang, Dengyin Zhang   +4 more
wiley   +1 more source

Gradient‐Aware Multiscale Feature Enhancement With Reliability Attention for Multi‐Focus Image Fusion

IET Image Processing, Volume 20, Issue 1, January/December 2026.
We propose MSFE‐Net, an algorithm for multi‐focus image fusion that combines gradient‐aware multi‐scale feature enhancement, focus‐reliability attention based on local‐variance statistics, and an enhanced spatial frequency fusion strategy using four‐directional gradients with local context and morphological refinement to produce robust decision maps ...
Haifeng Gong, Wen Liu, Yaofeng Wu, Xu Liang, Junhao Huang, Jun Xiao, Han Wang, Xiang Yan   +7 more
wiley   +1 more source

Robust Facial Recognition Using Unified Optimised Feature Fusion and Selection of Handcrafted and Deep Learning Features

IET Image Processing, Volume 20, Issue 1, January/December 2026.
This study presents a robust facial recognition framework based on a unified optimised feature vector that fuses handcrafted descriptors and deep learning embeddings. Using binary grey wolf optimisation for feature selection, the approach reduces redundancy while preserving discriminative power.
Farid Ayeche, Adel Alti
wiley   +1 more source