近年来随着无人值守井站在行业内逐渐普及,大大提高了采油巡查的效率,但无人值守井站也面临着一定的安全问题,不法分子为获取利益逃避追责,往往采用蒙面或者戴口罩的方式伪装后对采油设施进行破坏,传统的监控设施无法对不法分子进行准...近年来随着无人值守井站在行业内逐渐普及,大大提高了采油巡查的效率,但无人值守井站也面临着一定的安全问题,不法分子为获取利益逃避追责,往往采用蒙面或者戴口罩的方式伪装后对采油设施进行破坏,传统的监控设施无法对不法分子进行准确识别定位。为了消除这一安全隐患,笔者提出了基于YOLO(You Only Look Once)检测模型的人脸快速识别技术,解决嵌入式设备上人脸识别检测算法的检测精度低和检测速度慢的问题,实现了佩戴口罩情况下的人脸快速识别,为无人值守井站的安全提供了有力的保障,在一定程度上降低了石油公司的运营成本。展开更多
This study presents an innovative approach to improving the performance of YOLO-v8 model for small object detection in radar images.Initially,a local histogram equalization technique was applied to the original images...This study presents an innovative approach to improving the performance of YOLO-v8 model for small object detection in radar images.Initially,a local histogram equalization technique was applied to the original images,resulting in a notable enhancement in both contrast and detail representation.Subsequently,the YOLO-v8 backbone network was augmented by incorporating convolutional kernels based on a multidimensional attention mechanism and a parallel processing strategy,which facilitated more effective feature information fusion.At the model’s head,an upsampling layer was added,along with the fusion of outputs from the shallow network,and a detection head specifically tailored for small object detection,thereby further improving accuracy.Additionally,the loss function was modified to incorporate focal-intersection over union(IoU)in conjunction with scaled-IoU,which enhanced the model’s performance.A weighting strategy was also introduced,effectively improving detection accuracy for small targets.Experimental results demonstrate that the customized model outperforms traditional approaches across various evaluation metrics,including recall,precision,F1-score,and the receiver operating characteristic(ROC)curve,validating its efficacy and innovation in small object detection within radar imagery.The results indicate a substantial improvement in accuracy compared to conventional methods such as image segmentation and standard convolutional neural networks.展开更多
磁脉冲压接技术成形速度快、效率高,适合高强钢和铝、碳纤维等轻质材料的连接,在飞机工业中有广泛的应用前景。但目前针对磁脉冲压接管件的在线检测方法较少,不利于该技术实现自动化生产。针对磁脉冲压接管件压接质量的在线检测需求,提...磁脉冲压接技术成形速度快、效率高,适合高强钢和铝、碳纤维等轻质材料的连接,在飞机工业中有广泛的应用前景。但目前针对磁脉冲压接管件的在线检测方法较少,不利于该技术实现自动化生产。针对磁脉冲压接管件压接质量的在线检测需求,提出了一种基于改进YOLOv4–Tiny(You only look once v4–Tiny)检测网络和自适应图像处理的视觉检测方法。引入高效通道注意力(ECA)模块对YOLOv4–Tiny检测网络进行改进,基于自适应阈值分割算法和Canny边缘检测算法设计了一种自适应的压接深度提取算法,通过模拟工业生产环境采集了一批磁脉冲压接管件图像并划分为训练集和验证集,最后使用训练数据集对算法进行训练,并在验证集上验证训练得到的检测模型。结果表明,压接区域检测模型交并比阈值取0.5时的平均精确度(AP@0.5)为100%,交并比阈值分别取0.5、0.6、0.7、0.8时的平均精确度(AP@0.5:0.8)为93.14%,单帧运行时间为1.66ms;图像处理边缘提取算法平均偏差为0.85个像素,最大偏差为2.6个像素,单帧运行时间为3.49ms;完整压接深度提取算法平均偏差为0.313个像素,均方偏差为0.115平方像素,平均偏差率为1.35%,单帧运行时间为124.49ms。该算法能够在无辅助定位的条件下准确快速地实现磁脉冲压接工件压接深度提取,部署成本低,鲁棒性高,具有较高的应用价值。展开更多
文摘近年来随着无人值守井站在行业内逐渐普及,大大提高了采油巡查的效率,但无人值守井站也面临着一定的安全问题,不法分子为获取利益逃避追责,往往采用蒙面或者戴口罩的方式伪装后对采油设施进行破坏,传统的监控设施无法对不法分子进行准确识别定位。为了消除这一安全隐患,笔者提出了基于YOLO(You Only Look Once)检测模型的人脸快速识别技术,解决嵌入式设备上人脸识别检测算法的检测精度低和检测速度慢的问题,实现了佩戴口罩情况下的人脸快速识别,为无人值守井站的安全提供了有力的保障,在一定程度上降低了石油公司的运营成本。
基金supported by the Na‑tional Natural Science Foundation of China Joint Fund(No.U21B2028)the National Key R&D Program of China(No.2021YFC 2100100)the Shanghai Science and Technology Project(Nos.21JC1403400,23JC1402300).
文摘This study presents an innovative approach to improving the performance of YOLO-v8 model for small object detection in radar images.Initially,a local histogram equalization technique was applied to the original images,resulting in a notable enhancement in both contrast and detail representation.Subsequently,the YOLO-v8 backbone network was augmented by incorporating convolutional kernels based on a multidimensional attention mechanism and a parallel processing strategy,which facilitated more effective feature information fusion.At the model’s head,an upsampling layer was added,along with the fusion of outputs from the shallow network,and a detection head specifically tailored for small object detection,thereby further improving accuracy.Additionally,the loss function was modified to incorporate focal-intersection over union(IoU)in conjunction with scaled-IoU,which enhanced the model’s performance.A weighting strategy was also introduced,effectively improving detection accuracy for small targets.Experimental results demonstrate that the customized model outperforms traditional approaches across various evaluation metrics,including recall,precision,F1-score,and the receiver operating characteristic(ROC)curve,validating its efficacy and innovation in small object detection within radar imagery.The results indicate a substantial improvement in accuracy compared to conventional methods such as image segmentation and standard convolutional neural networks.
文摘磁脉冲压接技术成形速度快、效率高,适合高强钢和铝、碳纤维等轻质材料的连接,在飞机工业中有广泛的应用前景。但目前针对磁脉冲压接管件的在线检测方法较少,不利于该技术实现自动化生产。针对磁脉冲压接管件压接质量的在线检测需求,提出了一种基于改进YOLOv4–Tiny(You only look once v4–Tiny)检测网络和自适应图像处理的视觉检测方法。引入高效通道注意力(ECA)模块对YOLOv4–Tiny检测网络进行改进,基于自适应阈值分割算法和Canny边缘检测算法设计了一种自适应的压接深度提取算法,通过模拟工业生产环境采集了一批磁脉冲压接管件图像并划分为训练集和验证集,最后使用训练数据集对算法进行训练,并在验证集上验证训练得到的检测模型。结果表明,压接区域检测模型交并比阈值取0.5时的平均精确度(AP@0.5)为100%,交并比阈值分别取0.5、0.6、0.7、0.8时的平均精确度(AP@0.5:0.8)为93.14%,单帧运行时间为1.66ms;图像处理边缘提取算法平均偏差为0.85个像素,最大偏差为2.6个像素,单帧运行时间为3.49ms;完整压接深度提取算法平均偏差为0.313个像素,均方偏差为0.115平方像素,平均偏差率为1.35%,单帧运行时间为124.49ms。该算法能够在无辅助定位的条件下准确快速地实现磁脉冲压接工件压接深度提取,部署成本低,鲁棒性高,具有较高的应用价值。
