[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-662695-105":59,"doc-detail-662695-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","cyclegan-network-for-sheet-metal-welding-drawing-translation","Cyclegan Network for Sheet Metal Welding Drawing Translation","","Intelligent manufacturing depends on precise translation of engineering drawings, yet manual translation limits efficiency and can reduce manufacturing accuracy. This paper presents an automatic welding-structure drawing translation approach using CycleGAN for unpaired transfer learning. The generator and discriminator are built with U-Net and PatchGAN, while identity mapping is removed to improve grayscale noise robustness. Residual blocks and a sparse high-dimensional architecture enhance feature extraction and resolution. Experiments validate improved training stability and accuracy.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/cyclegan-network-for-sheet-metal-welding-drawing-translation/662695/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/cyclegan-network-for-sheet-metal-welding-drawing-translation/662695.png","ImageObject",300,407,{"name":92,"@type":93},"4398046744996","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-30","2026-10-28",true,{"@type":102,"interactionType":103,"userInteractionCount":19},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"Why is automatic translation of welding engineering drawings important in intelligent manufacturing?","Question",{"text":112,"@type":113},"Quality translation directly affects manufacturing accuracy, and manual translation reduces production efficiency. The paper targets higher accuracy with improved recognition efficiency.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What network structure does the paper use for welding drawing translation?",{"text":117,"@type":113},"It uses CycleGAN with a U-Net-based generator and a PatchGAN discriminator, learning feature mappings between real welding drawings and translated outputs.",{"name":119,"@type":110,"acceptedAnswer":120},"What improvements are proposed over the traditional CycleGAN model?",{"text":121,"@type":113},"The method removes color-sensitive identity mapping loss for grayscale drawings and adds residual blocks to the intermediate generator layers, aiming to improve feature extraction and generated resolution while maintaining training speed.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},662695,1793221757,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":66,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":133,"file_id":134,"file_url":135,"file_type":136,"file_size":137,"view_count":19,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":34,"language":138,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":139,"faqs":140,"seo_title":141,"seo_description":67,"update_tm":129,"read_time":142},4398046744996,"Cyclegan Network for Sheet Metal Welding Drawing Translation  \nZhiwei Songa, Hui Yaoa *, Dan Tiana, Gaohui Zhana  \na Xi'an Technological University, School of Mechatronic Engineering, 710000, Xi'an, Shaanxi, China.  \n[songwei9023@163.com](songwei9023@163.com);[yaohui@xatu.edu.cn](yaohui@xatu.edu.cn); [1394121483@qq.com](1394121483@qq.com); [1912740549@qq.com](1912740549@qq.com)  \nAbstract  \nIn intelligent manufacturing, the quality of machine translation engineering drawings will directly affect its manufacturing accuracy. Currently, most of the work is manually translated, greatly reducing production efficiency. This paper proposes an automatic translation method for welded structural engineering drawings based on Cyclic Generative Adversarial Networks (CycleGAN) . The CycleGAN network model of unpaired transfer learning is used to learn the feature mapping of real welding engineering drawings to realize automatic translation of engineering drawings. U-Net and PatchGAN are the main network for the generator and discriminator, respectively. Based on removing the identity mapping function, a high-dimensional sparse network is proposed to replace the traditional dense network for the Cyclegan generator to improve noise robustness. Increase the residual block hidden layer to increase theresolution of the generated graph. The improved and fine-tuned network models are experimentally validated, computing the gap between real and generated data. It meets the welding engineering precision standard and solves the main problem of low drawing recognition efficiency in the welding manufacturing process. The results show. After training with our model, the PSNR, SSIM and MSE of welding engineering drawings reach about 44.89%, 99.58% and 2.11, respectively, which are superior to traditional networks in both training speed and accuracy.  \nKeywords： Engineering Drawings; Deep Learning; CycleGAN; Image Translation; Sparse Algorithm  \n1. INTRODUCTION  \nGenerative Adversarial Net (GAN) networks were proposed by Goodfellow et al[1] . Inspired by zero-sum games in game theory, it treats the problem of generative recognition as a game of discriminator and generator. Generative adversarial networks consist of two deep learning networks, a generator, and a discriminator.  \nNowadays, the GAN network has been applied in 12 significant fields[2]， such as medicine, finance, geography, biology, etc. its network performance is relatively excellent, especially in remote sensing line recognition and style transfer; it has characteristics of high precision. Used the GAN network to identify and segment remote sensing images with high resolution and accuracy. In terms of digital line recognition, JianBo Guo, XiangRui Xu, XuDong Mao, HuiQing Qi[3][4][5][6]， and others used generative adversarial networks for recognition. Their recognition effect and accuracy requirements have a high guarantee. XuChen Zhen, Hua Xu and LeFei Zhang et al[7][8][9] .  \nFig 1. Two generators, Gx_y, Gy_x, and two discriminators, Dx, Dy. The real X inputs Gx_y, and the generator generates images in the Y domain. The resulting Y domain image is input to Gy_x, resulting in image X' . Compare X' and X loss. The discriminator is used to judge whether the data is true or false.  \nDue to the high precision and high resolution of generative adversarial networks, this paper proposes the idea of translating welding structural engineering drawings in the field of heavy industry equipment manufacturing based on the Cyclegan network.GAN network has more vital feature learning ability and generation ability than ordinary network model. Its training process is more complex and prone to gradient disappearance and model collapse[10] . GAN training instability problem, researchers, propose their solutions. Originally known as the improved DCGAN model [11], it relies on experimentally enumerating the discriminatorand generator architectures and finding a better network architecture setting, which ultima","cbCaiid2hnPs4oTM","https://ap.wps.com/l/cbCaiid2hnPs4oTM","pdf",1141199,"English","# 1. INTRODUCTION\n## GAN networks and applications\n## CycleGAN for unpaired translation\n## Contributions of the proposed method","[{\"question\":\"Why is automatic translation of welding engineering drawings important in intelligent manufacturing?\",\"answer\":\"Quality translation directly affects manufacturing accuracy, and manual translation reduces production efficiency. The paper targets higher accuracy with improved recognition efficiency.\"},{\"question\":\"What network structure does the paper use for welding drawing translation?\",\"answer\":\"It uses CycleGAN with a U-Net-based generator and a PatchGAN discriminator, learning feature mappings between real welding drawings and translated outputs.\"},{\"question\":\"What improvements are proposed over the traditional CycleGAN model?\",\"answer\":\"The method removes color-sensitive identity mapping loss for grayscale drawings and adds residual blocks to the intermediate generator layers, aiming to improve feature extraction and generated resolution while maintaining training speed.\"}]","Cyclegan Network for Sheet Metal Welding Drawing Translation | PDF",18]