[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122282-en":3,"doc-seo-122282-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},122282,1099514067438,"River Wang","https://ap-avatar.wpscdn.com/avatar/100002539ee87300030?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780474512215547542",6,"Technology","Training a Machine Learning Model to Read Illegible Barcodes","Illegible barcodes impede barcode use in real-world workflows, especially for direct part marking (DPM) where codes may be very small or have poor contrast such as dark-on-colored surfaces. The disclosure presents a supervised machine learning model trained on illegible barcode images paired with ground-truth legible barcodes or encoded information. Training compares model output to ground-truth, and the loss function updates model weights. In field use, a user photographs an illegible barcode, receives a legible version from the trained model, then decodes it with standard scanners to improve scanning performance in industrial and data-center settings.","Technical Disclosure Commons  \nDefensive Publications Series  \n28 Mar 2025  \nTraining a Machine Learning Model to Read Illegible Barcodes n/a  \nFollow this and additional works at: [https://www.tdcommons.org/dpubs_series](https://www.tdcommons.org/dpubs_series)  \nRecommended Citation  \nn/a, \"Training a Machine Learning Model to Read Illegible Barcodes\", Technical Disclosure Commons,(March 28, 2025)  \n[https://www.tdcommons.org/dpubs_series/7944](https://www.tdcommons.org/dpubs_series/7944)  \nThis work is licensed under a Creative Commons Attribution 4.0 License.  \nThis Article is brought to you for free and open access by Technical Disclosure Commons. It has been accepted for inclusion in Defensive Publications Series by an authorized administrator of Technical Disclosure Commons.  \nTraining a Machine Learning Model to Read Illegible Barcodes  \nABSTRACT  \nIllegible barcodes can hinder use of barcodes in many contexts. This document describes  \na machine learning (ML) model trained to read illegible barcodes. The training dataset used to  \ntrain the ML model comprises illegible barcodes (e.g., having low contrast, small size, wear-and  \ntear) and corresponding groundtruth. The ML model is trained to generate a legible version of  \nthe barcode. During training, the model output is compared with the groundtruth and the  \ndifference is fed to a loss function to adjust the weights ofthe model. In field use, a user can  \ncapture a photo of an illegible barcode and access the trained ML model to obtain a legible  \nversion. The described ML model can improve barcode scanning in industrial settings, data  \ncenters, and other contexts where barcodes are used.  \nKEYWORDS  \n● Illegible barcode  \n● Small barcode  \n● Barcode contrast  \n● Printed code  \n● Direct part marking (DPM)  \nBACKGROUND  \nBarcodes are used in many contexts to provide information in simple machine-readable format. For example, barcodes can be printed on paper and affixed to various objects, or can be directly printed on an object. In data centers and other industrial contexts, parts and equipment can be identified and tracked using barcodes that are laser printed directly on the various parts. This process, known as direct part marking (DPM) permanently marks parts with information to  \nPublished by Technical Disclosure Commons, 2025 2  \nenable tracking the parts throughout their life cycle. DPM uses machine-readable barcodes. In practice, DPM barcodes can be small in size. Further, DPM barcodes can have poor contrast with the rest ofthe part (e.g., dark green code on green part, instead of the more typical black-onwhite barcode label) . These characteristics can make DPM barcodes difficult to read. Image processing algorithms used by barcode scanners may not accurately read such small barcodes or barcodes with poor contrast.  \nDESCRIPTION  \nFig. 1: Training a machine learning model to read illegible barcodes  \nThis disclosure describes training a machine learning (ML) model to accurately decode  \nvisually illegible barcodes that are illegible for current techniques. Fig. 1 illustrates training a  \nmachine learning model to read illegible barcodes. A training dataset comprising illegible  \nbarcodes (102) with a variety of issues, e.g., low contrast, small size, wear-and-tear, etc. is  \n[https://www.tdcommons.org/dpubs_series/7944](https://www.tdcommons.org/dpubs_series/7944) 3  \nobtained. The training dataset also includes corresponding groundtruth (108), e.g., high-quality  \nbarcodes and/or the information encoded by each illegible barcode.  \nA machine learning model (104) is trained to generate model output that includes a legible version of the barcode (generated by the ML model based on the input. Using a model  \noutput readability detector (106), the model output is compared with the groundtruth for each  \nillegible barcode in the training dataset. For example, the model output barcode may be scanned  \nby barcode decoding software to obtain decoded information (“T","cbCainxZDuqshr5W","https://ap.wps.com/l/cbCainxZDuqshr5W","pdf",154496,1,5,"English","en",105,"# Abstract\n# Background\n# Description\n## Training workflow\n## Field use and deployment\n# Conclusion\n# References","[{\"question\":\"What problem does the document address?\",\"answer\":\"The document addresses difficulties scanning barcodes that are visually illegible, such as small codes, low-contrast prints, or codes affected by wear and tear.\"},{\"question\":\"How is the machine learning model trained?\",\"answer\":\"The model is trained with a dataset of illegible barcodes and corresponding ground-truth (high-quality barcodes and/or the encoded information). During training, the model output is compared against the ground-truth and the difference is used by a loss function to update model weights.\"},{\"question\":\"How can the trained model be used in the field?\",\"answer\":\"A user captures a photo of an illegible DPM barcode. The trained model outputs a legible version, which can then be decoded by regular barcode scanning software to retrieve the encoded information.\"}]","Training a Machine Learning Model to Read Illegible Barcodes | PDF",1785809806,13,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"training-a-machine-learning-model-to-read-illegible-barcodes","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/technology/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/training-a-machine-learning-model-to-read-illegible-barcodes/122282/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the document address?","Question",{"text":75,"@type":76},"The document addresses difficulties scanning barcodes that are visually illegible, such as small codes, low-contrast prints, or codes affected by wear and tear.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is the machine learning model trained?",{"text":80,"@type":76},"The model is trained with a dataset of illegible barcodes and corresponding ground-truth (high-quality barcodes and/or the encoded information). During training, the model output is compared against the ground-truth and the difference is used by a loss function to update model weights.",{"name":82,"@type":73,"acceptedAnswer":83},"How can the trained model be used in the field?",{"text":84,"@type":76},"A user captures a photo of an illegible DPM barcode. The trained model outputs a legible version, which can then be decoded by regular barcode scanning software to retrieve the encoded information.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,109,112,117,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},"Comic",60,"comic",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":110,"slug":111},50,"technology",{"id":113,"doc_module":4,"doc_module_name":46,"category_name":114,"show_sort_weight":115,"slug":116},7,"Healthcare",40,"healthcare",{"id":118,"doc_module":4,"doc_module_name":46,"category_name":119,"show_sort_weight":120,"slug":121},8,"Research & Report",30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":21,"slug":137},19,"General","general"]