Laser-directed energy deposition(L-DED)is an advanced additive manufacturing technology primarily adopted in metal three-dimensional printing systems.The L-DED process is characterized by various defects,thus necessit...Laser-directed energy deposition(L-DED)is an advanced additive manufacturing technology primarily adopted in metal three-dimensional printing systems.The L-DED process is characterized by various defects,thus necessitating the extensive use of in-situ monitoring to enable real-time adjustments of process parameters by detecting molten-pool features.To address the challenge of accurately extracting the molten-pool morphology from an undetached spatter,an innovative monitoring method based on the U-Net(U-shaped network)is proposed herein.A lightweight architecture accelerates the processing speed,whereas an enhanced loss function incorporating weight maps augments the segmentation precision.The model performance is evaluated by comparing its segmentation accuracy and processing speed with those of the conventional U-Net,using the mean intersection over union(MIoU)as the segmentation metric.The improved model demonstrates superior segmentation accuracy at the interface between the molten pool and spatter,with a peak MIoU of 0.9798 achieved on the test set.Furthermore,this model processes each image in an extremely short time of 17.9 ms.Using this segmentation algorithm,the error in extracting the molten-pool width from single-track experiments is within 0.1 mm.The proposed method for monitoring the molten-pool morphology is suitable for deployment in online monitoring systems,thus providing a foundation for subsequent process-parameter regulation.展开更多
The rapid pace of urban development has resulted in the widespread presence of construction equipment andincreasingly complex conditions in transmission corridors. These conditions pose a serious threat to the safeope...The rapid pace of urban development has resulted in the widespread presence of construction equipment andincreasingly complex conditions in transmission corridors. These conditions pose a serious threat to the safeoperation of the power grid.Machine vision technology, particularly object recognition technology, has beenwidelyemployed to identify foreign objects in transmission line images. Despite its wide application, the technique faceslimitations due to the complex environmental background and other auxiliary factors. To address these challenges,this study introduces an improved YOLOv8n. The traditional stepwise convolution and pooling layers are replacedwith a spatial-depth convolution (SPD-Conv) module, aiming to improve the algorithm’s efficacy in recognizinglow-resolution and small-size objects. The algorithm’s feature extraction network is improved by using a LargeSelective Kernel (LSK) attention mechanism, which enhances the ability to extract relevant features. Additionally,the SIoU Loss function is used instead of the Complete Intersection over Union (CIoU) Loss to facilitate fasterconvergence of the algorithm. Through experimental verification, the improved YOLOv8n model achieves adetection accuracy of 88.8% on the test set. The recognition accuracy of cranes is improved by 2.9%, which isa significant enhancement compared to the unimproved algorithm. This improvement effectively enhances theaccuracy of recognizing foreign objects on transmission lines and proves the effectiveness of the new algorithm.展开更多
The contact network dropper works in a harsh environment,and suffers from the impact effect of pantographs during running of trains,which may lead to faults such as slack and broken of the dropper wire and broken of t...The contact network dropper works in a harsh environment,and suffers from the impact effect of pantographs during running of trains,which may lead to faults such as slack and broken of the dropper wire and broken of the current-carrying ring.Due to the low intelligence and poor accuracy of the dropper fault detection network,an improved fully convolutional one-stage(FCOS)object detection network was proposed to improve the detection capability of the dropper condition.Firstly,by adjusting the parameterαin the network focus loss function,the problem of positive and negative sample imbalance in the network training process was eliminated.Secondly,the generalized intersection over union(GIoU)calculation was introduced to enhance the network’s ability to recognize the relative spatial positions of the prediction box and the bounding box during the regression calculation.Finally,the improved network was used to detect the status of dropper pictures.The detection speed was 150 sheets per millisecond,and the MAP of different status detection was 0.9512.Through the simulation comparison with other object detection networks,it was proved that the improved FCOS network had advantages in both detection time and accuracy,and could identify the state of dropper accurately.展开更多
Due to the complex environment of the university laboratory,personnel flow intensive,personnel irregular behavior is easy to cause security risks.Monitoring using mainstream detection algorithms suffers from low detec...Due to the complex environment of the university laboratory,personnel flow intensive,personnel irregular behavior is easy to cause security risks.Monitoring using mainstream detection algorithms suffers from low detection accuracy and slow speed.Therefore,the current management of personnel behavior mainly relies on institutional constraints,education and training,on-site supervision,etc.,which is time-consuming and ineffective.Given the above situation,this paper proposes an improved You Only Look Once version 7(YOLOv7)to achieve the purpose of quickly detecting irregular behaviors of laboratory personnel while ensuring high detection accuracy.First,to better capture the shape features of the target,deformable convolutional networks(DCN)is used in the backbone part of the model to replace the traditional convolution to improve the detection accuracy and speed.Second,to enhance the extraction of important features and suppress useless features,this paper proposes a new convolutional block attention module_efficient channel attention(CBAM_E)for embedding the neck network to improve the model’s ability to extract features from complex scenes.Finally,to reduce the influence of angle factor and bounding box regression accuracy,this paper proposes a newα-SCYLLA intersection over union(α-SIoU)instead of the complete intersection over union(CIoU),which improves the regression accuracy while increasing the convergence speed.Comparison experiments on public and homemade datasets show that the improved algorithm outperforms the original algorithm in all evaluation indexes,with an increase of 2.92%in the precision rate,4.14%in the recall rate,0.0356 in the weighted harmonic mean,3.60%in the mAP@0.5 value,and a reduction in the number of parameters and complexity.Compared with the mainstream algorithm,the improved algorithm has higher detection accuracy,faster convergence speed,and better actual recognition effect,indicating the effectiveness of the improved algorithm in this paper and its potential for practical application in laboratory scenarios.展开更多
针对部署于有限算力平台的YOLOv3(you only look once v3)算法对电容器外观缺陷存在检测速度较慢的问题,提出了基于YOLOv3算法改进的轻量化算法MQYOLOv3。首先采用轻量化网络MobileNet v2作为特征提取模块,通过利用深度可分离式卷积替...针对部署于有限算力平台的YOLOv3(you only look once v3)算法对电容器外观缺陷存在检测速度较慢的问题,提出了基于YOLOv3算法改进的轻量化算法MQYOLOv3。首先采用轻量化网络MobileNet v2作为特征提取模块,通过利用深度可分离式卷积替换一般卷积操作,使得模型的参数量大幅度降低进而提高模型的检测速度,同时也带来了检测精度的降低;然后在网络结构中嵌入空间金字塔池化结构实现局部特征与全局特征的融合、引入距离交并比(distance intersection over union,DIoU)损失函数优化交并比(intersection over union,IoU)损失函数以及使用Mish激活函数优化Leaky ReLU激活函数来提高模型的检测精度。本文采用自制的电容器外观缺陷数据集进行实验,轻量化MQYOLOv3算法的平均精度均值(mean average precision,mAP)为87.96%,较优化前降低了1.16%,检测速度从1.5 FPS提升到7.7 FPS。实验表明,本文设计的轻量化MQYOLOv3算法在保证检测精度的同时,提高了检测速度。展开更多
基金supported by National Natural Science Foundation of China(Grant Nos.52305440,52204263)Natural Science Foundation of Changsha City(Grant Nos.kq2208272,kq2208274)+1 种基金Tribology Science Fund of the State Key Laboratory of Tribology in Advanced Equipment(Grant SKLTKF22B09)National Key Research and Development Program of China(2022YFB3706902).
文摘Laser-directed energy deposition(L-DED)is an advanced additive manufacturing technology primarily adopted in metal three-dimensional printing systems.The L-DED process is characterized by various defects,thus necessitating the extensive use of in-situ monitoring to enable real-time adjustments of process parameters by detecting molten-pool features.To address the challenge of accurately extracting the molten-pool morphology from an undetached spatter,an innovative monitoring method based on the U-Net(U-shaped network)is proposed herein.A lightweight architecture accelerates the processing speed,whereas an enhanced loss function incorporating weight maps augments the segmentation precision.The model performance is evaluated by comparing its segmentation accuracy and processing speed with those of the conventional U-Net,using the mean intersection over union(MIoU)as the segmentation metric.The improved model demonstrates superior segmentation accuracy at the interface between the molten pool and spatter,with a peak MIoU of 0.9798 achieved on the test set.Furthermore,this model processes each image in an extremely short time of 17.9 ms.Using this segmentation algorithm,the error in extracting the molten-pool width from single-track experiments is within 0.1 mm.The proposed method for monitoring the molten-pool morphology is suitable for deployment in online monitoring systems,thus providing a foundation for subsequent process-parameter regulation.
基金the Natural Science Foundation of Shandong Province(ZR2021QE289)State Key Laboratory of Electrical Insulation and Power Equipment(EIPE22201).
文摘The rapid pace of urban development has resulted in the widespread presence of construction equipment andincreasingly complex conditions in transmission corridors. These conditions pose a serious threat to the safeoperation of the power grid.Machine vision technology, particularly object recognition technology, has beenwidelyemployed to identify foreign objects in transmission line images. Despite its wide application, the technique faceslimitations due to the complex environmental background and other auxiliary factors. To address these challenges,this study introduces an improved YOLOv8n. The traditional stepwise convolution and pooling layers are replacedwith a spatial-depth convolution (SPD-Conv) module, aiming to improve the algorithm’s efficacy in recognizinglow-resolution and small-size objects. The algorithm’s feature extraction network is improved by using a LargeSelective Kernel (LSK) attention mechanism, which enhances the ability to extract relevant features. Additionally,the SIoU Loss function is used instead of the Complete Intersection over Union (CIoU) Loss to facilitate fasterconvergence of the algorithm. Through experimental verification, the improved YOLOv8n model achieves adetection accuracy of 88.8% on the test set. The recognition accuracy of cranes is improved by 2.9%, which isa significant enhancement compared to the unimproved algorithm. This improvement effectively enhances theaccuracy of recognizing foreign objects on transmission lines and proves the effectiveness of the new algorithm.
基金supported by Natural Science Foundation of Gansu Province(No.20JR10RA216)。
文摘The contact network dropper works in a harsh environment,and suffers from the impact effect of pantographs during running of trains,which may lead to faults such as slack and broken of the dropper wire and broken of the current-carrying ring.Due to the low intelligence and poor accuracy of the dropper fault detection network,an improved fully convolutional one-stage(FCOS)object detection network was proposed to improve the detection capability of the dropper condition.Firstly,by adjusting the parameterαin the network focus loss function,the problem of positive and negative sample imbalance in the network training process was eliminated.Secondly,the generalized intersection over union(GIoU)calculation was introduced to enhance the network’s ability to recognize the relative spatial positions of the prediction box and the bounding box during the regression calculation.Finally,the improved network was used to detect the status of dropper pictures.The detection speed was 150 sheets per millisecond,and the MAP of different status detection was 0.9512.Through the simulation comparison with other object detection networks,it was proved that the improved FCOS network had advantages in both detection time and accuracy,and could identify the state of dropper accurately.
基金This study was supported by the National Natural Science Foundation of China(No.61861007)Guizhou ProvincialDepartment of Education Innovative Group Project(QianJiaohe KY[2021]012)Guizhou Science and Technology Plan Project(Guizhou Science Support[2023]General 412).
文摘Due to the complex environment of the university laboratory,personnel flow intensive,personnel irregular behavior is easy to cause security risks.Monitoring using mainstream detection algorithms suffers from low detection accuracy and slow speed.Therefore,the current management of personnel behavior mainly relies on institutional constraints,education and training,on-site supervision,etc.,which is time-consuming and ineffective.Given the above situation,this paper proposes an improved You Only Look Once version 7(YOLOv7)to achieve the purpose of quickly detecting irregular behaviors of laboratory personnel while ensuring high detection accuracy.First,to better capture the shape features of the target,deformable convolutional networks(DCN)is used in the backbone part of the model to replace the traditional convolution to improve the detection accuracy and speed.Second,to enhance the extraction of important features and suppress useless features,this paper proposes a new convolutional block attention module_efficient channel attention(CBAM_E)for embedding the neck network to improve the model’s ability to extract features from complex scenes.Finally,to reduce the influence of angle factor and bounding box regression accuracy,this paper proposes a newα-SCYLLA intersection over union(α-SIoU)instead of the complete intersection over union(CIoU),which improves the regression accuracy while increasing the convergence speed.Comparison experiments on public and homemade datasets show that the improved algorithm outperforms the original algorithm in all evaluation indexes,with an increase of 2.92%in the precision rate,4.14%in the recall rate,0.0356 in the weighted harmonic mean,3.60%in the mAP@0.5 value,and a reduction in the number of parameters and complexity.Compared with the mainstream algorithm,the improved algorithm has higher detection accuracy,faster convergence speed,and better actual recognition effect,indicating the effectiveness of the improved algorithm in this paper and its potential for practical application in laboratory scenarios.
文摘针对部署于有限算力平台的YOLOv3(you only look once v3)算法对电容器外观缺陷存在检测速度较慢的问题,提出了基于YOLOv3算法改进的轻量化算法MQYOLOv3。首先采用轻量化网络MobileNet v2作为特征提取模块,通过利用深度可分离式卷积替换一般卷积操作,使得模型的参数量大幅度降低进而提高模型的检测速度,同时也带来了检测精度的降低;然后在网络结构中嵌入空间金字塔池化结构实现局部特征与全局特征的融合、引入距离交并比(distance intersection over union,DIoU)损失函数优化交并比(intersection over union,IoU)损失函数以及使用Mish激活函数优化Leaky ReLU激活函数来提高模型的检测精度。本文采用自制的电容器外观缺陷数据集进行实验,轻量化MQYOLOv3算法的平均精度均值(mean average precision,mAP)为87.96%,较优化前降低了1.16%,检测速度从1.5 FPS提升到7.7 FPS。实验表明,本文设计的轻量化MQYOLOv3算法在保证检测精度的同时,提高了检测速度。
