[1]季倓正,邵允学,吕摇 刚.基于深度学习的钢坯入炉前定位技术[J].计算机技术与发展,2022,32(S2):36-40.[doi:10. 3969 / j. issn. 1673-629X. 2022. S2. 006]
　JI Tan-zheng,SHAO Yun-xue,LYU Gang.Positioning Technology of Billet before Furnace Based on Deep Learning[J].,2022,32(S2):36-40.[doi:10. 3969 / j. issn. 1673-629X. 2022. S2. 006]
点击复制

基于深度学习的钢坯入炉前定位技术()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年S2期
页码:
36-40
栏目:
人工智能
出版日期:
2022-12-11

文章信息/Info

Title:
Positioning Technology of Billet before Furnace Based on Deep Learning
文章编号:
1673-629X(2022)S2-0036-05
作者:
季倓正1 ; 邵允学1 ; 吕摇 刚2
1. 南京工业大学 计算机科学与技术学院,江苏 南京 211816;
2. 上海策立工程技术有限公司,上海 201900
Author(s):
JI Tan-zheng1 ; SHAO Yun-xue1 ; LYU Gang2
1. School of Computer Science and Technology,Nanjing Tech University,Nanjing 211816,China;
2. Shanghai Celi Engineering Technology Co. ,Ltd. ,Shanghai 201900,China
关键词:
炉前定位; 图像分割; ResNet; U-Net; 横向分割网络
Keywords:
furnace front positioning; image segmentation; ResNet; U-Net; horizontal partition network
分类号:
TP391. 41
DOI:
10. 3969 / j. issn. 1673-629X. 2022. S2. 006
摘要:
随着深度学习技术的不断发展,深度学习与传统工业生产流程相融合成为近些年研究和应用的热点。 在传统钢铁生产过程中,钢坯入炉位置是人工根据钢坯的横向位置是否正确来确认和控制的,存在控制不精确、生产效率低、人工成本高等不足之处。 针对以上问题,文中通过机器视觉方法实现钢坯在炉前位置的实时精确定位,提出了一种基于钢坯横向位置分割的炉前定位方法,该方法以 U-Net 神经网络为基础结构,用 ResNet-34 网络的标准残差块对原 U-Net 网络进行特征提取与特征融合,这样既保留了 ResNet 的残差结构以加深网络层次,也使这两种网络有效的融合起来。 经实验,横向分割网络的像素精度可以达到 99. 7% ,而标准的 ResNet 网络仅达到 92. 2% ,达到了现场应用精度要求。
Abstract:
With the continuous development of deep learning technology, the integration of deep learning and traditional industrialproduction process has become a research and application hotspot in recent years. In the traditional steel production process,the billetfeeding position is manually confirmed and controlled according to whether the transverse position of the billet is correct. There are someshortcomings,such as inaccurate control,low production efficiency and high labor cost. To solve the above problems,we realize the real-time accurate positioning of billet in front of the furnace by machine vision method and propose a furnace positioning method based onbillet transverse position segmentation. This method takes u-net neural network as the basic structure,and uses the standard residual blockof resnet-34 network to extract and fuse the features of the original u-net network. It not only retains the residual structure of RESNETto deepen the network level, but also makes the two networks integrate effectively. Through experiments, the pixel accuracy of thehorizontal segmentation network can reach 99. 7% , while the standard RESNET network can only reach 92. 2% , which meets theaccuracy requirements of field application,and lays a foundation for the automation and intelligence of billet production process.

相似文献/References:

[1]蒋璐璐 王适 王宝成 李慧敏 李鑫慧.一种改进的标记分水岭遥感图像分割方法[J].计算机技术与发展,2010,(01):36.
　JIANG Lu-lu,WANG Shi,WANG Bao-cheng,et al.Segmentation of Remote Sensing Image Based on an Improved Labeling Watershed Algorithm[J].,2010,(S2):36.
[2]张少娴 俞琼.基于时空相关性预测的运动估计的优化[J].计算机技术与发展,2010,(01):100.
　ZHANG Shao-xian,YU Qiong.An Optimization Method for Spatiotemporal Predictive Motion Estimation[J].,2010,(S2):100.
[3]王兴 冯子亮.基于自适应初始值的FCM聚类图像分割[J].计算机技术与发展,2010,(03):101.
　WANG Xing,FENG Zi-liang.An Image Segmentation Algorithm Based on Adaptive Initialization FCM Clustering[J].,2010,(S2):101.
[4]何小娜 逄焕利.基于二维直方图和改进蚁群聚类的图像分割[J].计算机技术与发展,2010,(03):128.
　HE Xiao-na,PANG Huan-li.Image Segmentation Based on Improved Ant Colony Clustering and Two- Dimensional Histogram[J].,2010,(S2):128.
[5]宋淑娜 李金霞 胡学坤 高尚.一种自适应模糊阈值区间的图像分割方法[J].计算机技术与发展,2010,(05):121.
　SONG Shu-na,LI Jin-xia,HU Xue-kun,et al.A Method of Adaptive Fuzzy Threshold Region for Image Segmentation[J].,2010,(S2):121.
[6]来磊 卢文科 邓开连.基于二维Tsallis交叉熵直线型图像阈值分割方法[J].计算机技术与发展,2010,(06):105.
　LAI Lei,LU Wen-ke,DENG Kai-lian.New Image Thresholding Segmentation Methods Based on Two-Dimensional Tsallis Cross-Entropy Liner-Type[J].,2010,(S2):105.
[7]黄长专 王彪 杨忠.图像分割方法研究[J].计算机技术与发展,2009,(06):76.
　HUANG Chang-zhuan,WANG Biao,YANG Zhong.A Study on Image Segmentation Techniques[J].,2009,(S2):76.
[8]李光耀 聂诗良.基于小波分解和模糊聚类的图像分割方法[J].计算机技术与发展,2009,(06):121.
　LI Guang-yao,NIE Shi-liang.Image Segment Algorithm Based on Wavelet Decomposition and Fuzzy Clustering Theory[J].,2009,(S2):121.
[9]吴亚 汪继文.水平集图像分割中重新初始化规避的探索[J].计算机技术与发展,2009,(09):69.
　WU Ya,WANG Ji-wen.Avoidance of Re- Initialization in Level Set Image Segmentation[J].,2009,(S2):69.
[10]李鑫环 陈立潮 赵红艳 赵勇.基于多小波分析与SOFM的MR图像分割算法研究[J].计算机技术与发展,2009,(09):104.
　LI Xin-huan,CHEN Li-chao,ZHAO Hong-yan,et al.Research on MR Image Segmentation Based on Multi- wavelet Analysis and SOFM[J].,2009,(S2):104.

更新日期/Last Update: 2022-10-10