Aiming at the problem of low surface defect detection accuracy of industrial products, an object detection method based on simplified spatial pyramid pooling fast(Sim SPPF) hybrid pooling improved you only look once v...Aiming at the problem of low surface defect detection accuracy of industrial products, an object detection method based on simplified spatial pyramid pooling fast(Sim SPPF) hybrid pooling improved you only look once version 5s(YOLOV5s) model is proposed. The algorithm introduces channel attention(CA) module, simplified SPPF feature vector pyramid and efficient intersection over union(EIOU) loss function. Feature vector pyramids fuse high-dimensional and low-dimensional features, which makes semantic information richer. The CA mechanism performs maximum pooling and average pooling operations on the feature map. Hybrid pooling comprehensively improves detection computing efficiency and accurate deployment ability. The results show that the improved YOLOV5s model is better than the original YOLOV5s model. The average test accuracy(mAP) can reach 91.8%, which can be increased by 17.4%, and the detection speed can reach 108 FPS, which can be increased by 18 FPS. The improved model is practicable, and the overall performance is better than other conventional models.展开更多
针对现有目标跟踪算法存在目标感知能力较弱、实时性不足以及目标易丢失的问题,提出一种结合金字塔池化和目标运动轨迹的单目标跟踪算法(Pyramid pooling Transformer for single object tracking,PPTTrack)。首先,算法采用PVT作为骨干...针对现有目标跟踪算法存在目标感知能力较弱、实时性不足以及目标易丢失的问题,提出一种结合金字塔池化和目标运动轨迹的单目标跟踪算法(Pyramid pooling Transformer for single object tracking,PPTTrack)。首先,算法采用PVT作为骨干网络构建深度神经网络,建立模板图和搜索图之间的双向信息流通道,提升跟踪算法的目标感知能力;其次,在骨干网络中引入金字塔池化方法缩短K,V的序列长度,降低模型复杂度,提高算法运行速率;最后,将跟踪目标的运动轨迹构建为一个Motion Token输入到编码器中进行特征融合,利用运动轨迹连续性预测目标位置,以解决目标丢失问题。实验结果表明,上述算法在GOT-10K、LaSOT、UAV123以及TrackingNet四个基准数据集上的性能均达到了较为先进的水平。展开更多
针对目前人工巡视导致的变电站设备及生产行为异常检测效率低、人工风险高等问题,提出改进你只看一次11纳米版(you only look once version 11 nano, YOLOv11n)模型。首先,通过设计基于自注意力机制的3尺度卷积双路径可变核(convolution...针对目前人工巡视导致的变电站设备及生产行为异常检测效率低、人工风险高等问题,提出改进你只看一次11纳米版(you only look once version 11 nano, YOLOv11n)模型。首先,通过设计基于自注意力机制的3尺度卷积双路径可变核(convolutional three-scale kernel-adaptive dual-path self-attention mechanism, C3k2-SA)模块在较小特征图衔接特征融合部分,优化了网络结构,增强了全局特征提取能力。然后,在主干网络末层引入了基于注意力机制的特征增强(feature enhancement, FEN)模块,动态调整不同区域的特征权重,实现自适应的特征增强,缓解深层网络中的梯度消失问题。最后,对拼接(concatenate, Concat)模块进行优化,通过卷积层调整通道数,采用池化和sigmoid激活函数进行特征的精细处理,提高了模型对不同类型特征的自适应,增强了特征融合效果,同时抑制了无关或冗余特征,防止过拟合。结果表明,与原始YOLOv11n模型相比,改进YOLOv11n模型的精确率、召回率、平均精确率均值分别上升了1.7、6.6、3.6个百分点。改进YOLOv11n模型能够提高变电站异常状态检测的准确性,为智能变电站的异常检测工作提供一定参考。展开更多
With the appearance of Internet promotes GIS on both technical aspect and applied aspect,traditional GIS encounters huge blocks on Internet platform. In the field of computer ,Distributed Computing technology has seen...With the appearance of Internet promotes GIS on both technical aspect and applied aspect,traditional GIS encounters huge blocks on Internet platform. In the field of computer ,Distributed Computing technology has seen fast progress with three industrial standards. And on GIS field,openGIS Consortium has drawn a series of specifications,which,combining with the three standards,provides enterprise GIS application with foundamental facilities. The authors firstly illustrate the architectures with three models-thin clent,medium client and thick client,then present a mixed model which transfers both vector and raster data between client side and server side with the advantage of high interactivity in clinet and medium data transfer. Afterward,the authors give two enterprise schema based on two standards :COM+and EJB. Finally,the authors list some key technologies which will affect the future WebGIS-openGIS,XML, Windows. net, CORBA, EJB,etc.展开更多
基金supported by the Tianjin Postgraduate Research Innovation Project (No.2022SKY286)the National Science and the National Key Research and Development Program (No.2022YFF0706000)。
文摘Aiming at the problem of low surface defect detection accuracy of industrial products, an object detection method based on simplified spatial pyramid pooling fast(Sim SPPF) hybrid pooling improved you only look once version 5s(YOLOV5s) model is proposed. The algorithm introduces channel attention(CA) module, simplified SPPF feature vector pyramid and efficient intersection over union(EIOU) loss function. Feature vector pyramids fuse high-dimensional and low-dimensional features, which makes semantic information richer. The CA mechanism performs maximum pooling and average pooling operations on the feature map. Hybrid pooling comprehensively improves detection computing efficiency and accurate deployment ability. The results show that the improved YOLOV5s model is better than the original YOLOV5s model. The average test accuracy(mAP) can reach 91.8%, which can be increased by 17.4%, and the detection speed can reach 108 FPS, which can be increased by 18 FPS. The improved model is practicable, and the overall performance is better than other conventional models.
文摘针对水下环境复杂性带来的多尺度目标检测挑战,提出了改进算法WPS-YOLOv8。设计了小波池化卷积模块(wavelet pooling convolution,WPConv),该模块通过小波池化技术降低通道压缩后特征图的分辨率,有效抑制了传统下采样过程中产生的频率混叠伪影,提升了特征提取质量和表达能力。提出了局部逐点分组重排卷积模块(partial pointwise group shuffle convolution,PGConv),该模块通过结合局部卷积与逐点卷积,能够在减少信息冗余的同时保持通道间的信息交互,弥补了深度可分离卷积的不足,增强了特征融合效果。提出了ShapeLoss损失函数,综合考虑影响不同尺度目标检测精度的因素,通过集成Shape-IoU和Shape-NWD两种损失测度,有效提升了对多尺度目标的总体检测精度。实验结果显示,相较于YOLOv8,WPS-YOLOv8在URPC2018和UTDAC2020水下数据集上的平均精度均值(mean average precision,mAP)分别提升了8.6和4.4个百分点,展现了其在水下多尺度目标检测中的出色表现。
文摘针对现有目标跟踪算法存在目标感知能力较弱、实时性不足以及目标易丢失的问题,提出一种结合金字塔池化和目标运动轨迹的单目标跟踪算法(Pyramid pooling Transformer for single object tracking,PPTTrack)。首先,算法采用PVT作为骨干网络构建深度神经网络,建立模板图和搜索图之间的双向信息流通道,提升跟踪算法的目标感知能力;其次,在骨干网络中引入金字塔池化方法缩短K,V的序列长度,降低模型复杂度,提高算法运行速率;最后,将跟踪目标的运动轨迹构建为一个Motion Token输入到编码器中进行特征融合,利用运动轨迹连续性预测目标位置,以解决目标丢失问题。实验结果表明,上述算法在GOT-10K、LaSOT、UAV123以及TrackingNet四个基准数据集上的性能均达到了较为先进的水平。
文摘With the appearance of Internet promotes GIS on both technical aspect and applied aspect,traditional GIS encounters huge blocks on Internet platform. In the field of computer ,Distributed Computing technology has seen fast progress with three industrial standards. And on GIS field,openGIS Consortium has drawn a series of specifications,which,combining with the three standards,provides enterprise GIS application with foundamental facilities. The authors firstly illustrate the architectures with three models-thin clent,medium client and thick client,then present a mixed model which transfers both vector and raster data between client side and server side with the advantage of high interactivity in clinet and medium data transfer. Afterward,the authors give two enterprise schema based on two standards :COM+and EJB. Finally,the authors list some key technologies which will affect the future WebGIS-openGIS,XML, Windows. net, CORBA, EJB,etc.
