一种基于CNN深度学习的焊接机器人视觉模型
|
摘要
为了准确地识别复杂环境下的焊缝目标,建立了一种基于深度学习的焊接机器人视觉模型,该模型采用局部联接和全联接相结合的CNN卷积神经网络结构,局部联接由3个卷积层和子采样层交替组成,用于焊接目标的特征提取,全连接层由输入层、隐层和输出层组成,作为分类器用于焊缝目标识别.采样了一千多幅焊接目标图像样本用于CNN的网络训练,分析了不同CNN网络结构参数对模型的影响.结果表明,该视觉模型对焊接目标的平移、旋转和比例缩放表现出良好的鲁棒性,可以应用到焊接机器人的视觉导航.
In order to accurately recognize the weld target in complex environment, a visual model of welding robot based on deep learning was established. The model adoped a convolutional neural network(CNN) combining local connection and full connection. The local connection was composed of 3 convolution layers(C) and 3 subsampling layers(S) with C-S alternating mode for feature extraction of welding target. The full connection layer was composed of input layer,hidden layer and output layer as a classifier for weld target recognition. More than 1 000 image samples of welding targets were sampled for CNN network training, and the influence of different CNN structure parameters on the model was analyzed. The test results show that the visual model was robust to the translation, rotation and scaling of welding targets, and could be applied to the visual navigation of welding robots.
In order to accurately recognize the weld target in complex environment, a visual model of welding robot based on deep learning was established. The model adoped a convolutional neural network(CNN) combining local connection and full connection. The local connection was composed of 3 convolution layers(C) and 3 subsampling layers(S) with C-S alternating mode for feature extraction of welding target. The full connection layer was composed of input layer,hidden layer and output layer as a classifier for weld target recognition. More than 1 000 image samples of welding targets were sampled for CNN network training, and the influence of different CNN structure parameters on the model was analyzed. The test results show that the visual model was robust to the translation, rotation and scaling of welding targets, and could be applied to the visual navigation of welding robots.
引文
[1]陈希章,陈善本.弧焊机器人起始焊接位置图像识别与定位[J].焊接学报,2009,30(4):17-20.Chen Xizhang,Chen Shanben.Recognition and positioning of start welding position for arc welding robot[J].Transactions of the China Welding Institution,2009,30(4):17-20.
[2]叶艳辉,张华,潘继銮,等.大型构件水下焊接机器人系统[J].焊接学报,2015,36(11):41-44.Ye Yanhui,Zhang Hua,Pan Jiluan,et al.System of underwater welding robot for large-scale structure[J].Transactions of the China Welding Institution,2015,36(11):41-44.
[3]杨雪君,许燕玲,黄色吉,等.一种基于结构光的V型坡口焊缝特征点识别算法[J].上海交通大学学报,2016,50(10):1573-1577.Yang Xuejun,Xu Yanling,Huang Seji,et al.A recognition algorithm for feature points of V groove welds based on structured light[J].Journal of Shanghai Jiaotong University,2016,50(10):1573-1577.
[4]郭亮,张华.狭小空间直角角焊缝识别跟踪焊接系统[J].焊接学报,2017,38(11):21-26.Guo Liang,Zhang Hua.Identification tracking welding system for right angle fillet seam in narrow space[J].Transactions of the China Welding Institution,2017,38(11):21-26.
[5]丛焕武,郭福娟,吕飞,等.基于CCD图像处理的焊缝识别技术研究[J].电子测量技术,2012,35(3):73-76.Cong Huanwu,Guo Fujuan,Lu Fei,et al.Research on weld seam recognition technology based on CCD image processing[J].Electronic Measurement Technology,2012,35(3):73-76.
[6]Hinton G E,Osindero S,Teh Y W.A fast learning algorithm for deep belief nets[J].Neural Computation,2006,18(7):1527-1554.
[7]Hinton G E,Salakhutdinov R R.Reducing the dimensionality of data with neural networks[J].Science,2006,313(5786):504-507.
[8]Lecun Y,Bengio Y,Hinton G E.Deep learning[J].Nature,2015,521(7553):436-444.
[9]Simonyan K,Zisserman A.Very deep convolutional networks for large-scale image recognition[C]//Proc.Of International Conference on Learning Representations(ICLR),San Diego,CA,USA,2015:1-14.
[10]Yim J,Ju J,Jung H.Image classification using convolutional neural networks with multi-stage feature[C]//Proc.of 3rd International Conference on Robot Intelligence Technology and Applications,Springer,Switzerland,2015:587-593.
[11]Tompson J,Goroshin R,Jain A,et al.Efficient object localization using convolutional networks[C]//Proc.of IEEE Conference on Computer Vision and Pattern Recognition(CVPR),Boston,MA,USA,2015:648-656.
[2]叶艳辉,张华,潘继銮,等.大型构件水下焊接机器人系统[J].焊接学报,2015,36(11):41-44.Ye Yanhui,Zhang Hua,Pan Jiluan,et al.System of underwater welding robot for large-scale structure[J].Transactions of the China Welding Institution,2015,36(11):41-44.
[3]杨雪君,许燕玲,黄色吉,等.一种基于结构光的V型坡口焊缝特征点识别算法[J].上海交通大学学报,2016,50(10):1573-1577.Yang Xuejun,Xu Yanling,Huang Seji,et al.A recognition algorithm for feature points of V groove welds based on structured light[J].Journal of Shanghai Jiaotong University,2016,50(10):1573-1577.
[4]郭亮,张华.狭小空间直角角焊缝识别跟踪焊接系统[J].焊接学报,2017,38(11):21-26.Guo Liang,Zhang Hua.Identification tracking welding system for right angle fillet seam in narrow space[J].Transactions of the China Welding Institution,2017,38(11):21-26.
[5]丛焕武,郭福娟,吕飞,等.基于CCD图像处理的焊缝识别技术研究[J].电子测量技术,2012,35(3):73-76.Cong Huanwu,Guo Fujuan,Lu Fei,et al.Research on weld seam recognition technology based on CCD image processing[J].Electronic Measurement Technology,2012,35(3):73-76.
[6]Hinton G E,Osindero S,Teh Y W.A fast learning algorithm for deep belief nets[J].Neural Computation,2006,18(7):1527-1554.
[7]Hinton G E,Salakhutdinov R R.Reducing the dimensionality of data with neural networks[J].Science,2006,313(5786):504-507.
[8]Lecun Y,Bengio Y,Hinton G E.Deep learning[J].Nature,2015,521(7553):436-444.
[9]Simonyan K,Zisserman A.Very deep convolutional networks for large-scale image recognition[C]//Proc.Of International Conference on Learning Representations(ICLR),San Diego,CA,USA,2015:1-14.
[10]Yim J,Ju J,Jung H.Image classification using convolutional neural networks with multi-stage feature[C]//Proc.of 3rd International Conference on Robot Intelligence Technology and Applications,Springer,Switzerland,2015:587-593.
[11]Tompson J,Goroshin R,Jain A,et al.Efficient object localization using convolutional networks[C]//Proc.of IEEE Conference on Computer Vision and Pattern Recognition(CVPR),Boston,MA,USA,2015:648-656.