期刊文献+
任意字段
题名或关键词
题名
关键词
摘要
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息
共找到306篇文章
< 1 2 … 16 >
每页显示 20 50 100
已选0条
引用分析
参考文献
引证文献
 统计分析
检索结果
已选文献
显示方式：
ECANet:Enhanced Convolutional Attention Network for Liver Segmentation 认领 引用
1
作者 Yuyan Ning Haiyun Huang +3 位作者 Legend Zhang Wei Wei Hao Quan Bo Yang 《Computer Modeling in Engineering & Sciences》 SCIE EI 2026年第6期1155-1172,共18页
Hybrid CNN-Transformer models are widely used in medical image segmentation because they combine CNN-based local feature extraction with Transformer-based global context modeling.Despite their popularity,these models ... Hybrid CNN-Transformer models are widely used in medical image segmentation because they combine CNN-based local feature extraction with Transformer-based global context modeling.Despite their popularity,these models face several challenges,including computational complexity,noise blurring,and information loss.This paper proposes an enhanced convolutional attention network(ECANet)for liver segmentation.ECANet uses a U-shaped architecture with efficient channel-attention-based skip connections.Both the encoder and decoder are constructed using enhanced convolutional Transformer(ECT)blocks,where group convolution is integrated into the convolutional attention module for efficient Token embedding and channel disentanglement,and a Token-wise multi-layer perceptron(MLP)branch is incorporated into the wide-focus module to improve feature representation across channels.Deep supervision and a hybrid of Binary Cross-Entropy(BCE)and Dice loss are used to improve boundary accuracy.We evaluate the proposed model on the publicly available LiTS17 dataset.Experiments show that ECANet outperforms the compared CNN-based and CNN-Transformer baseline models on both quantitative and qualitative measures. 展开更多
关键词 Medical image segmentation convolutional transformer group convolution efficient channel attention deep supervision
暂未订购 下载PDF
Grid Side Distributed Energy Storage Cloud Group End Region Hierarchical Time-Sharing Configuration Algorithm Based onMulti-Scale and Multi Feature Convolution Neural Network 认领 引用 被引量：1
2
作者 Wen Long Bin Zhu +3 位作者 Huaizheng Li Yan Zhu Zhiqiang Chen Gang Cheng 《Energy Engineering》 EI 2023年第5期1253-1269,共17页
There is instability in the distributed energy storage cloud group end region on the power grid side.In order to avoid large-scale fluctuating charging and discharging in the power grid environment and make the capaci... There is instability in the distributed energy storage cloud group end region on the power grid side.In order to avoid large-scale fluctuating charging and discharging in the power grid environment and make the capacitor components showa continuous and stable charging and discharging state,a hierarchical time-sharing configuration algorithm of distributed energy storage cloud group end region on the power grid side based on multi-scale and multi feature convolution neural network is proposed.Firstly,a voltage stability analysis model based onmulti-scale and multi feature convolution neural network is constructed,and the multi-scale and multi feature convolution neural network is optimized based on Self-OrganizingMaps(SOM)algorithm to analyze the voltage stability of the cloud group end region of distributed energy storage on the grid side under the framework of credibility.According to the optimal scheduling objectives and network size,the distributed robust optimal configuration control model is solved under the framework of coordinated optimal scheduling at multiple time scales;Finally,the time series characteristics of regional power grid load and distributed generation are analyzed.According to the regional hierarchical time-sharing configuration model of“cloud”,“group”and“end”layer,the grid side distributed energy storage cloud group end regional hierarchical time-sharing configuration algorithm is realized.The experimental results show that after applying this algorithm,the best grid side distributed energy storage configuration scheme can be determined,and the stability of grid side distributed energy storage cloud group end region layered timesharing configuration can be improved. 展开更多
关键词 Multiscale and multi feature convolution neural network distributed energy storage at grid side cloud group end region layered time-sharing configuration algorithm
暂未订购 下载PDF
Convolutional neural network based data interpretable framework for Alzheimer’s treatment planning 认领 引用
3
作者 Sazia Parvin Sonia Farhana Nimmy Md Sarwar Kamal 《Visual Computing for Industry,Biomedicine,and Art》 EI 2024年第1期375-386,共12页
Alzheimer’s disease(AD)is a neurological disorder that predominantly affects the brain.In the coming years,it is expected to spread rapidly,with limited progress in diagnostic techniques.Various machine learning(ML)a... Alzheimer’s disease(AD)is a neurological disorder that predominantly affects the brain.In the coming years,it is expected to spread rapidly,with limited progress in diagnostic techniques.Various machine learning(ML)and artificial intelligence(AI)algorithms have been employed to detect AD using single-modality data.However,recent developments in ML have enabled the application of these methods to multiple data sources and input modalities for AD prediction.In this study,we developed a framework that utilizes multimodal data(tabular data,magnetic resonance imaging(MRI)images,and genetic information)to classify AD.As part of the pre-processing phase,we generated a knowledge graph from the tabular data and MRI images.We employed graph neural networks for knowledge graph creation,and region-based convolutional neural network approach for image-to-knowledge graph generation.Additionally,we integrated various explainable AI(XAI)techniques to interpret and elucidate the prediction outcomes derived from multimodal data.Layer-wise relevance propagation was used to explain the layer-wise outcomes in the MRI images.We also incorporated submodular pick local interpretable model-agnostic explanations to interpret the decision-making process based on the tabular data provided.Genetic expression values play a crucial role in AD analysis.We used a graphical gene tree to identify genes associated with the disease.Moreover,a dashboard was designed to display XAI outcomes,enabling experts and medical professionals to easily comprehend the predic-tion results. 展开更多
关键词 Multimodal Region-based convolutional neural network Layer-wise relevance propagation Submodular pick local interpretable model-agnostic explanations Graphical genes tree Alzheimer’s disease
暂未订购 下载PDF
Branch-Activated Multi-Domain Convolutional Neural Network for Visual Tracking 认领 引用 被引量：2
4
作者 CHEN Yimin LU Rongron +1 位作者 ZOU Yibo ZHANG Yanhui 《Journal of Shanghai Jiaotong university(Science)》 EI 2018年第3期360-367,共8页
Convolutional neural networks （CNNs） have been applied in state-of-the-art visual tracking tasks to represent the target. However, most existing algorithms treat visual tracking as an object-specific task. Therefore... Convolutional neural networks （CNNs） have been applied in state-of-the-art visual tracking tasks to represent the target. However, most existing algorithms treat visual tracking as an object-specific task. Therefore, the model needs to be retrained for different test video sequences. We propose a branch-activated multi-domain convolutional neural network （BAMDCNN）. In contrast to most existing trackers based on CNNs which require frequent online training, BAMDCNN only needs offine training and online fine-tuning. Specifically, BAMDCNN exploits category-specific features that are more robust against variations. To allow for learning category-specific information, we introduce a group algorithm and a branch activation method. Experimental results on challenging benchmark show that the proposed algorithm outperforms other state-of-the-art methods. What＇s more, compared with CNN based trackers, BAMDCNN increases tracking speed. 展开更多
关键词 convolutional neural network(CNN) category-specific feature group algorithm branch activation method
暂未订购 下载PDF
面向遥感图像超分辨率重建的高效通道注意力算法 认领 引用 被引量：1
5
作者 陈晓璇 仝晓丹 +2 位作者 李耀维 蔡誉涵 姜博 《西北大学学报（自然科学版）》 CAS CSCD 北大核心 2026年第1期108-117,共10页
遥感成像设备普遍面临距离远、成像分辨率低的问题,直接影响遥感图像质量和应用效果。针对这一问题,提出一种基于高效通道注意力的特征增强超分辨率重建网络模型,将图像超分辨率重建技术引入到遥感图像处理领域,利用分组卷积特征增强模... 遥感成像设备普遍面临距离远、成像分辨率低的问题,直接影响遥感图像质量和应用效果。针对这一问题,提出一种基于高效通道注意力的特征增强超分辨率重建网络模型,将图像超分辨率重建技术引入到遥感图像处理领域,利用分组卷积特征增强模块对图像进行特征提取和增强,然后利用高效通道注意力和非对称卷积并联构成的注意力模块,建立起图像不同区域之间的相互关系,重建出高分辨率图像。实验结果表明,该算法在WHU-RS19测试集上的峰值信噪比和结构相似性分别为28.70 dB、0.7539,分别比次优方法提高了0.19 dB和0.0066,重建图像的细节也更加丰富,从客观指标和主观视觉上都验证了该算法的有效性。 展开更多
关键词 遥感图像 超分辨率 分组卷积 通道注意力
暂未订购 下载PDF
A MCG-GFAM-MRDCM Model for Accurate Building Electricity Load Forecasting 认领 引用
6
作者 Chuan Lin Weixian Chen Guangtao Hao 《Energy Engineering》 EI 2026年第8期208-237,共30页
Accurate building electricity load forecasting(BELF)can provide a regulatory basis for building energy management systems and promote the transition of buildings toward low-carbon and intelligent operation modes.Howev... Accurate building electricity load forecasting(BELF)can provide a regulatory basis for building energy management systems and promote the transition of buildings toward low-carbon and intelligent operation modes.However,building electricity load is influenced by historical loads,as well as outside environmental conditions such as humidity and temperature,which reduces the prediction accuracy of models.To tackle these challenges,this study presents a BELF model,which consists of a modal component grouping approach,grouped feature attention mechanism,and multi-scale residual depthwise convolution memory module.First,the modal component grouping method analyzes building electricity load in the time domain,frequency domain(via fast fourier transform,FFT),and complexity(via sample entropy,SE),and then performs clustering to achieve precise decomposition of load components with different fluctuation characteristics.Second,the grouped feature attention mechanism assigns suitable importance to various input features to emphasize key factors affecting prediction accuracy.Third,the multi-scale residual depthwise convolution memory module mitigates the impact of long and short-term load variations on BELF by employing residual blocks of depthwise convolution layers with different kernel sizes.Meanwhile,gated recurrent units are used to identify the time-dependent trends of building load.Experimental results on public buildings show that the proposed model outperforms existing models,achieving more than 2.4%improvement in MAPE prediction performance. 展开更多
关键词 Building electricity load forecasting modal component grouping method depthwise convolutional neural network attention mechanism
暂未订购 下载PDF
基于自适应群体掩码图卷积网络的行人轨迹预测 认领 引用
7
作者 陈旺兴 桑海峰 刘晴 《电子学报》 EI CAS CSCD 北大核心 2026年第2期774-784,共11页
行人轨迹预测对于提升自动驾驶和服务机器人的决策能力以及降低未来潜在碰撞风险方面具有重要意义。然而,由于行人群体内外社会交互关系的差异性和复杂性,现有研究往往未能对群内和群外交互关系进行显式区分与独立建模,不同类型交互特... 行人轨迹预测对于提升自动驾驶和服务机器人的决策能力以及降低未来潜在碰撞风险方面具有重要意义。然而,由于行人群体内外社会交互关系的差异性和复杂性,现有研究往往未能对群内和群外交互关系进行显式区分与独立建模,不同类型交互特征在模型学习过程中相互混淆,难以精准刻画行人在复杂场景下的真实运动模式,进而制约了模型预测性能的进一步提升。因此,本文提出了一种基于自适应群体掩码图卷积网络(Adaptive Group Masked Graph Convolution Network,AGMGCN)的行人轨迹预测模型,通过对群内和群外交互关系进行独立建模,从而提升模型轨迹预测的准确性。该模型首先构建社会图并采用自注意力机制进行处理,以获得注意力矩阵用于初步表示行人之间的交互关系。后续设计了时频域卷积模块,通过在时域和频域同时对注意力矩阵进一步处理,生成用于表征行人时空交互关系的时频交互矩阵,以实现对行人复杂动态交互更准确的刻画。为有效区分并独立建模群内和群外交互,模型设计了自适应群体掩码模块,根据行人之间的特征相似性自适应确定阈值,并通过阈值处理生成群内掩码矩阵和群外掩码矩阵,为后续群内和群外交互关系的独立建模提供支持。在此基础上,将时频交互矩阵与群外和群内掩码矩阵相结合,并分别应用图卷积捕捉群内交互特征和群外交互特征,从而实现行人群内和群外交互关系的独立建模。最后,模型设计了特征融合模块动态加权融合群内交互特征和群外交互特征,并通过时间卷积网络完成行人未来轨迹的预测。在ETH、UCY和SDD数据集上的实验结果表明,在仅使用23.9 K模型参数的条件下,本文提出的方法相较于DSTIGCN在平均位移误差和终点位移误差上分别降低了12%和20%,验证了所提方法在预测精度和计算成本方面的优势。 展开更多
关键词 行人轨迹预测 自适应群体掩码 图卷积网络 时频域卷积模块 时间卷积网络
暂未订购 下载PDF
轻量化的ResNet50图像分类模型 认领 引用 被引量：1
8
作者 王鑫 王在顺 《计算机与现代化》 2026年第3期88-94,共7页
针对残差网络在处理图像分类任务时,可能因特征提取能力不足而影响准确率的问题,本文提出一种EResNext50-EFPN图像分类模型。首先,设计分组卷积注意力残差单元,该单元在残差块的主干路径上采用分组卷积替代传统卷积,并融入ECA注意力机制... 针对残差网络在处理图像分类任务时,可能因特征提取能力不足而影响准确率的问题,本文提出一种EResNext50-EFPN图像分类模型。首先,设计分组卷积注意力残差单元,该单元在残差块的主干路径上采用分组卷积替代传统卷积,并融入ECA注意力机制,不仅增强了模型的表达能力,还有效地降低了参数量和计算量;其次,本文对下采样模块进行优化处理;最后,设计一个多特征融合模块,该模块能够有效地融合来自不同层级的特征,从而进一步提升图像分类的准确率。在模型训练方面,采用预热策略与余弦退火衰减方法相结合的方式,以确保模型能够更稳定地收敛。实验结果表明,与原始的ResNet50模型相比,本文提出的EResNext50-EFPN模型在CIFAR-100数据集上的分类精度提高了2.95个百分点,而参数数量却仅为ResNet50模型的69%,计算量缩减至ResNet50模型的60%。 展开更多
关键词 图像分类 特征提取 分组卷积 注意力机制 多特征融合
暂未订购 下载PDF
改进YOLOv8n算法的船舶工业钢材表面缺陷检测 认领 引用
9
作者 刘鹏 侯博文 +2 位作者 王彩霞 姜晓娇 丛海芳 《兵工学报》 EI CAS CSCD 北大核心 2026年第3期35-49,共15页
为提高船舶工业中钢材表面缺陷检测的准确性,针对现有YOLOv8n算法在特征提取能力不足、检测精度低以及特征融合不充分等问题,提出一种基于改进YOLOv8n的钢材表面缺陷检测方法。构建高效视觉空间金字塔池化增强层聚合网络(Efficient Visi... 为提高船舶工业中钢材表面缺陷检测的准确性,针对现有YOLOv8n算法在特征提取能力不足、检测精度低以及特征融合不充分等问题,提出一种基于改进YOLOv8n的钢材表面缺陷检测方法。构建高效视觉空间金字塔池化增强层聚合网络(Efficient Vision Transformer-Spatial Pyramid Pooling with Enhanced Layer Aggregation Network,EfficientViT-SPPELAN),以增强多维度特征提取能力;设计多尺度时空卷积(Multi-Scale Spatial-Temporal Convolution,MSSTConv)实现多尺度特征融合;在此基础上构建多尺度时空(Multi-Scale Spatial-Temporal,MSST)模块以获取丰富的上下文信息,提高缺陷定位精度并降低计算复杂度,从而提升算法的推理效率。基于东北大学表面缺陷数据集(Northeastern University Surface Defect Dataset,NEU-DET)和镀锌钢10类缺陷检测数据集(Galvanized Steel 10-category Defect Detection Dataset,GC10-DET)两个数据集的实验结果表明,所提方法的检测精准度相较于原始YOLOv8n算法分别提升6.8%和5.7%,均值平均精确率mAP@0.5分别提高3.7%和7.9%;每秒帧数(Frames Per Second,FPS)分别达到189帧/s和142帧/s。研究结果表明,该方法在提升检测精度的同时保持较高计算效率,能够有效完成船舶钢材表面缺陷的定位和类别识别,满足工业场景对检测精度与实时性的需求。 展开更多
关键词 缺陷检测 YOLOv8n算法 多尺度时空模块 多尺度时空卷积 分组注意力
暂未订购 下载PDF
基于自适应融合CNN—OF特征和LSTM网络的猪攻击行为识别 认领 引用
10
作者 陈晨 孙博 +3 位作者 Juan Steibel Janice Siegford 韩俊杰 Tomas Norton 《中国农机化学报》 北大核心 2026年第2期275-282,共8页
为识别群养猪攻击行为,提出一种基于自适应融合CNN—OF特征和LSTM网络的算法。在两个猪栏中每栏混养8头猪3天,每天收集8 h的视频作为数据集。从猪栏1的3天视频中标记出1200个攻击1 s片段和1200个非攻击1 s片段,选择80%的片段作为训练集... 为识别群养猪攻击行为,提出一种基于自适应融合CNN—OF特征和LSTM网络的算法。在两个猪栏中每栏混养8头猪3天,每天收集8 h的视频作为数据集。从猪栏1的3天视频中标记出1200个攻击1 s片段和1200个非攻击1 s片段,选择80%的片段作为训练集,其余20%作为验证集。从猪栏2的3天视频中标记出1254个攻击1 s片段和85146个非攻击1 s片段作为测试集。首先,采用Horn—Schunck(HS)方法计算光流(OF)的大小和方向角,并根据CNN特征图的维度划分光流方向角的范围。然后,在每个方向角范围内统计光流大小的直方图,通过空间维度变换将直方图转化为特征图。最后,通过权重叠加将此特征图与CNN特征图进行自适应融合并输入LSTM网络以识别攻击。采用VGG16—OF—LSTM、ResNet50—OF—LSTM、InceptionV3—OF—LSTM和Xception—OF—LSTM算法识别猪攻击行为的准确率分别为97.5%、97.8%、98.7%、99.3%。结果表明,CNN—OF—SLTM算法能够识别猪攻击行为。提出的自适应特征融合方法CNN—OF具有一定通用性。 展开更多
关键词 群养猪 攻击识别 卷积神经网络 光流 自适应融合 长短期记忆
暂未订购 下载PDF
基于高阶递归网络的单幅图像去雨滴模型 认领 引用
11
作者 包玉刚 贾皓翔 赵旦峰 《系统工程与电子技术》 EI CSCD 北大核心 2026年第1期12-21,共10页
目前单幅图像去雨滴模型提取大尺度雨滴特征的能力较差,导致精度不高,无法很好地应用在复杂多变的实际场景中。为此,提出一种基于高阶递归网络的单幅图像去雨滴模型。首先,利用结合注意力机制的分组卷积构建一种双尺度注意力残差模块,... 目前单幅图像去雨滴模型提取大尺度雨滴特征的能力较差,导致精度不高,无法很好地应用在复杂多变的实际场景中。为此,提出一种基于高阶递归网络的单幅图像去雨滴模型。首先,利用结合注意力机制的分组卷积构建一种双尺度注意力残差模块,更好地提取大尺度雨滴的有效特征。其次,设计一种高阶递归特征传递机制,有效强化了这些特征从局部到整体的传递作用。最后,提出一种双尺度残差门控循环单元,建立了对递归计算中逐阶段特征的反馈过程,进一步提高了模型的性能。实验结果表明,提出的高阶递归网络在公开的基准数据集上取得了当前最优的性能表现,较好解决了当前算法精度不足的问题。 展开更多
关键词 高阶递归 深度学习 单幅图像去雨滴 双尺度残差 分组卷积 门控循环单元
暂未订购 下载PDF
基于改进YOLOv12的多场景老人跌倒检测 认领 引用
12
作者 王建霞 田楠楠 +1 位作者 李璇 张晓明 《河北工业科技》 CAS 2026年第4期319-327,共9页
为了解决多场景老人跌倒检测中模型计算参数量大与复杂场景精度不足的问题,提出一种基于改进YOLOv12的多场景老人跌倒检测算法。首先,在YOLOv12主干网络嵌入分组混洗卷积(grouped shuffle convolution,GSConv),同时引入Triplet注意力机... 为了解决多场景老人跌倒检测中模型计算参数量大与复杂场景精度不足的问题,提出一种基于改进YOLOv12的多场景老人跌倒检测算法。首先,在YOLOv12主干网络嵌入分组混洗卷积(grouped shuffle convolution,GSConv),同时引入Triplet注意力机制,并将颈部结构替换为轻量化颈部结构Slim Neck;其次,通过自建跌倒数据集对改进YOLOv12模型进行训练与验证;最后,在测试集上进行检测实验,评估改进模型的检测精度与鲁棒性。结果表明:改进后的模型相比原YOLOv12模型,参数量从2.56 MB降低至2.34 MB,减少了约8.59%;FLOPs从6.3 G降低至5.4 G,减少了约14.29%;mAP@50从83.8%提升至87.6%,mAP@50-95从51.3%提升至55.4%,Recall从76.5%提升至81.3%。改进后的YOLOv12模型实现了轻量化与检测精度的有效平衡,可为多场景老人跌倒检测提供一种有效的解决方案。 展开更多
关键词 计算机图像处理 YOLOv12 跌倒检测 Slim Neck 分组混洗卷积 Triplet注意力机制
暂未订购 下载PDF
基于多尺度特征增强的改进YOLOv11n桑蚕健康状态实时检测研究 认领 引用
13
作者 易云飞 史英杰 +2 位作者 陈旺 施丽媛 史翔宇 《电子技术应用》 2026年第7期142-150,共9页
为解决传统桑蚕养殖中病蚕识别滞后、检测精度不足的问题,提出一种改进YOLOv11n的桑蚕健康状态检测方法(MSDA-YOLO)。该方法的核心创新包括:(1)在主干网络中引入多尺度分组空洞卷积(Multi-scale Grouped Dilated Convolution,MSGDC)与C... 为解决传统桑蚕养殖中病蚕识别滞后、检测精度不足的问题,提出一种改进YOLOv11n的桑蚕健康状态检测方法(MSDA-YOLO)。该方法的核心创新包括:(1)在主干网络中引入多尺度分组空洞卷积(Multi-scale Grouped Dilated Convolution,MSGDC)与C3k2结合设计了C3k2MS模块,以提升对不同尺度特征的提取能力;(2)首次在桑蚕检测研究中引入DySample动态上采样器,提高在密集场景下对蚕体的空间细节感知与几何结构保持能力;(3)使用自适应空间特征融合策略(Adaptively Spatial Feature Fusion,ASFF)改造YOLOv11的检测头,优化多尺度特征融合效能,抑制尺度间特征冲突。实验结果表明,MSDA-YOLO在多项评估指标上均表现出色。具体而言,该方法在精确率上达到87.0%,召回率为77.8%,mAP@0.5为87.3%,mAP@0.5:0.95为52.8%,均领先于基线模型YOLOv11n和目前主流的YOLO系列检测模型,在桑蚕健康检测方面具有明显优势。 展开更多
关键词 YOLOv11n 多尺度分组空洞卷积 DySample 自适应空间特征融合
暂未订购 下载PDF
基于改进YOLO11的苹果品质分类检测模型 认领 引用 被引量：1
14
作者 刘宗东 何辉波 +2 位作者 李华英 黄云 刘亚伦 《农机化研究》 北大核心 2026年第8期179-187,共9页
苹果果实表面特征复杂、纹理多样,现有方法难以对其进行精确检测与分类,无法满足实际需求。为此,基于YOLO11网络框架构建了YOLO11-FCFL网络模型。首先,提出了一种新的特征增强扩展金字塔网络(Feature Enhancement Expansion Pyramid Net... 苹果果实表面特征复杂、纹理多样,现有方法难以对其进行精确检测与分类,无法满足实际需求。为此,基于YOLO11网络框架构建了YOLO11-FCFL网络模型。首先,提出了一种新的特征增强扩展金字塔网络(Feature Enhancement Expansion Pyramid Network,FEEPN),通过多层特征的融合与扩散,提升模型对多尺度信息的捕获能力;其次,在主干网络中引入多样化分支块(Deep Diverse Branch Block)改进C3k2模块,形成C3k2-DeepDBB模块,增强主干网络的特征提取能力;再次,设计了一种轻量化组卷积检测头(Light Group Head),替换原有的解耦头,降低模型复杂度并提高检测效率;最后,利用焦点调制网络(Focal Nets)的焦点调制模块(Focal Modulation)取代快速空间金字塔池化模块,实现更丰富的上下文聚合与交互。在自建苹果数据集上进行分类检测试验,结果表明:改进后的YOLO11-FCFE模型准确率、召回率、平均精度均值mAP@0.5和mAP@0.5-0.95分别达到了87.5%、88.1%、93.0%、90.9%,较原有模型分别提高了0.5、5.3、2.6、2.4个百分点。与DINO、Faster R-CNN、RetinaNet、YOLOv5、YOLOv8、YOLOv10模型相比,mAP@0.5分别提高了0.7、2.6、3.2、2.4、2.0、4.3个百分点。YOLO11-FCFL模型为复杂表面特征的苹果识别与分类提供了高效解决方案,对推动苹果分类自动化具有重要意义。 展开更多
关键词 苹果检测 YOLO11 品质分类 特征金字塔 多样化分支块 焦点调制 组卷积
暂未订购 下载PDF
基于YOLOv8的PCB缺陷检测改进算法 认领 引用 被引量：1
15
作者 刘爽 吕俊良 +2 位作者 秦宇航 秦丹丹 孙佳慧 《吉林大学学报（理学版）》 CAS 北大核心 2026年第2期344-350,共7页
针对工业印刷电路板缺陷检测任务中,小目标特征不明显且检测精度不足的问题,提出一种基于YOLOv8算法的改进算法.首先,通过增删特征图尺寸以适应印刷电路板缺陷检测,并借鉴加权双向特征金字塔网络结构保留原始图像的特征;其次,利用分组... 针对工业印刷电路板缺陷检测任务中,小目标特征不明显且检测精度不足的问题,提出一种基于YOLOv8算法的改进算法.首先,通过增删特征图尺寸以适应印刷电路板缺陷检测,并借鉴加权双向特征金字塔网络结构保留原始图像的特征;其次,利用分组卷积在颈部设计一个轻量化模块进行特征提取,提高检测精度的同时降低了模型复杂度;最后,在小目标检测头前引入可增强特征表现能力的坐标注意力模块,进一步提高检验精度.实验结果表明,改进后的算法能将检测精度mAP@0.5提升至95.4%,并使检测速度FPS(帧每秒)达到105.4,可以更好地满足工业检测对精度和实时性的要求. 展开更多
关键词 印刷电路板缺陷检测 神经网络 注意力机制 分组卷积
暂未订购 下载PDF
Video Super-Resolution via Effective Spatio-Temporal Alignment Network 认领 引用
16
作者 Bin Guo Xin Wang +5 位作者 Hao Wen Yuhong Fu Jinxing Li Hui Ma Haoqian Wang Yong Xu 《CAAI Transactions on Intelligence Technology》 SCIE EI CSCD 2026年第3期726-738,共13页
Extracting spatio-temporal cues from neighbouring frames is challenging in video super-resolution(VSR).Although deformable alignment-based VSR methods have shown promise in aligning neighbouring frames with the refere... Extracting spatio-temporal cues from neighbouring frames is challenging in video super-resolution(VSR).Although deformable alignment-based VSR methods have shown promise in aligning neighbouring frames with the reference frame,most existing methods rely on one or a few traditional convolutions to estimate motion offsets for spatio-temporal alignment,restricting receptive field size and alignment accuracy.To address these limitations,we propose an effective spatio-temporal alignment network(ESTA-Net)for VSR.The core component of our method is the group convolution-based alignment module(GCBAM),which utilises cascaded group convolutions to learn offsets across both the original and downsampled resolutions.By employing group convolutions rather than traditional convolutions,GCBAM enables the deformable alignment to achieve a wider receptive field with lower computational cost,thereby improving the accuracy of offset estimation.Additionally,the bi-scale alignment strategy within GCBAM enhances robustness to complex and large-scale motions.Furthermore,we introduce an attention-based feature enhancement module(AFEM)to refine the aligned features,focusing on critical details to improve reconstruction quality.Extensive experiments on standard benchmarks show that our ESTA-Net achieves superior VSR performance against other advanced methods,while maintaining a good equilibrium between model size and performance. 展开更多
关键词 group convolution motion offsets spatio‐temporal alignment video super‐resolution
暂未订购 下载PDF
FMCSNet: Mobile Devices-Oriented Lightweight Multi-Scale Object Detection via Fast Multi-Scale Channel Shuffling Network Model 认领 引用
17
作者 Lijuan Huang Xianyi Liu +1 位作者 Jinping Liu Pengfei Xu 《Computers, Materials & Continua》 SCIE EI 2026年第1期1292-1311,共20页
The ubiquity of mobile devices has driven advancements in mobile object detection.However,challenges in multi-scale object detection in open,complex environments persist due to limited computational resources.Traditio... The ubiquity of mobile devices has driven advancements in mobile object detection.However,challenges in multi-scale object detection in open,complex environments persist due to limited computational resources.Traditional approaches like network compression,quantization,and lightweight design often sacrifice accuracy or feature representation robustness.This article introduces the Fast Multi-scale Channel Shuffling Network(FMCSNet),a novel lightweight detection model optimized for mobile devices.FMCSNet integrates a fully convolutional Multilayer Perceptron(MLP)module,offering global perception without significantly increasing parameters,effectively bridging the gap between CNNs and Vision Transformers.FMCSNet achieves a delicate balance between computation and accuracy mainly by two key modules:the ShiftMLP module,including a shift operation and an MLP module,and a Partial group Convolutional(PGConv)module,reducing computation while enhancing information exchange between channels.With a computational complexity of 1.4G FLOPs and 1.3M parameters,FMCSNet outperforms CNN-based and DWConv-based ShuffleNetv2 by 1%and 4.5%mAP on the Pascal VOC 2007 dataset,respectively.Additionally,FMCSNet achieves a mAP of 30.0(0.5:0.95 IoU threshold)with only 2.5G FLOPs and 2.0M parameters.It achieves 32 FPS on low-performance i5-series CPUs,meeting real-time detection requirements.The versatility of the PGConv module’s adaptability across scenarios further highlights FMCSNet as a promising solution for real-time mobile object detection. 展开更多
关键词 Object detection lightweight network partial group convolution multilayer perceptron
暂未订购 下载PDF
TX-GGCA:a lightweight model based on Tiny-Xception for predicting axillary lymph node metastasis 认领 引用
18
作者 Wenyuan ZENG Fengnian LIU +3 位作者 Miduo TAN Yibo ZHANG Jing LONG Lin TANG 《Optoelectronics Letters》 EI 2026年第5期302-308,共7页
Many studies based on convolutional neural networks(CNNs)for breast cancer axillary lymph node(ALN)images have focused on large sample analysis and clinical parameter integration,while limited attention has been paid ... Many studies based on convolutional neural networks(CNNs)for breast cancer axillary lymph node(ALN)images have focused on large sample analysis and clinical parameter integration,while limited attention has been paid to lightweight models for small ALN datasets.In this paper,we have selected a small number of ALN ultrasound image datasets as the research subject and designed a TX-GGCA model,consisting of the Tiny-Xception model and the global grouping coordinate attention(GGCA).The TX-GGCA demonstrated an accuracy of 99.14%and an area under curve(AUC)of 0.9997 in classifying normal and abnormal ALN images,outperforming the best traditional model(accuracy:95.69%,AUC:0.9932).It showed the potential value of this model for clinical diagnosis in primary hospitals with limited sample sizes. 展开更多
关键词 ultrasound image datasets clinical parameter integrationwhile convolutional neural networks cnns lightweight models global grouping coordinate attention ggca large sample analysis tiny xception global grouping coordinate attention
暂未订购 下载PDF
基于分组交叉门控机制图卷积神经网络的源荷超短期功率联合预测方法 认领 引用 被引量：1
19
作者 赵婉冰 杨强 陈源奕 《高电压技术》 EI CAS CSCD 北大核心 2026年第2期639-651,I0005-I0007,共13页
多场站多类型源荷功率联合预测场景中,传统的时间序列预测模型难以高效提取多类型源荷的时空相关特征和内在隐含联系,导致功率预测精度不足。针对该问题,该文提出一种融合改进图卷积神经网络和分组交叉门控机制的源荷超短期功率联合预... 多场站多类型源荷功率联合预测场景中,传统的时间序列预测模型难以高效提取多类型源荷的时空相关特征和内在隐含联系,导致功率预测精度不足。针对该问题,该文提出一种融合改进图卷积神经网络和分组交叉门控机制的源荷超短期功率联合预测方法。首先,构建了基于改进型时空图卷积神经网络的源荷超短期功率联合预测模型,可充分提取源侧(风电、光伏)与荷侧(电负荷、热负荷)功率的时空关联特征。进而,设计了一种分组交叉门控机制并集成到改进的图卷积网络中,可使不同类型源荷信息得以交叉调制,有效利用了源荷之间的内在隐含联系,从而显著提升了多场站多类型源荷超短期功率联合预测的精度。最后,基于我国北部某地区的风光电源和电热负荷数据进行了对比实验,结果验证了所提方法的有效性和优越性。 展开更多
关键词 超短期功率预测 功率联合预测 图卷积网络 分组交叉门控机制 时空相关性
暂未订购 下载PDF
Effects of various information scenarios on layer-wise relevance propagation-based interpretable convolutional neural networks for air handling unit fault diagnosis 认领 引用 被引量：1
20
作者 Chenglong Xiong Guannan Li +3 位作者 Ying Yan Hanyuan Zhang Chengliang Xu Liang Chen 《Building Simulation》 SCIE EI CSCD 2024年第10期1709-1730,共22页
Deep learning(DL),especially convolutional neural networks(CNNs),has been widely applied in air handling unit(AHU)fault diagnosis(FD).However,its application faces two major challenges.Firstly,the accessibility of ope... Deep learning(DL),especially convolutional neural networks(CNNs),has been widely applied in air handling unit(AHU)fault diagnosis(FD).However,its application faces two major challenges.Firstly,the accessibility of operational state variables for AHU systems is limited in practical,and the effectiveness and applicability of existing DL methods for diagnosis require further validation.Secondly,the interpretability performance of DL models under various information scenarios needs further exploration.To address these challenges,this study utilized publicly available ASHRAE RP-1312 AHU fault data and employed CNNs to construct three FD models under three various information scenarios.Furthermore,the layer-wise relevance propagation(LRP)method was used to interpret and explain the effects of these three various information scenarios on the CNN models.An R-threshold was proposed to systematically differentiate diagnostic criteria,which further elucidates the intrinsic reasons behind correct and incorrect decisions made by the models.The results showed that the CNN-based diagnostic models demonstrated good applicability under the three various information scenarios,with an average diagnostic accuracy of 98.55%.The LRP method provided good interpretation and explanation for understanding the decision mechanism of CNN models for the unlimited information scenarios.For the very limited information scenario,since the variables are restricted,although LRP can reveal key variables in the model’s decision-making process,these key variables have certain limitations in terms of data and physical explanations for further improving the model’s interpretation.Finally,an in-depth analysis of model parameters—such as the number of convolutional layers,learning rate,βparameters,and training set size—was conducted to examine their impact on the interpretative results.This study contributes to clarifying the effects of various information scenarios on the diagnostic performance and interpretability of LRP-based CNN models for AHU FD,which helps provide improved reliability of DL models in practical applications. 展开更多
关键词 air handling unit(AHU) fault diagnosis convolutional neural network(CNN) layer-wise relevance propagation(LRP) interpretation and explanation various information scenarios
已选0条
引用分析
参考文献
引证文献
 统计分析
检索结果
已选文献
上一页 1 2 … 16 下一页 到第 页
在线咨询 使用帮助 返回顶部 意见反馈