[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86333-en":3,"doc-seo-86333-105":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":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86333,7971461741311,"Ophelia","https://ap-avatar.wpscdn.com/avatar/74000253aff267980c6?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779345379180704826",8,"Research & Report","MicroCharNet: Less is More for License Plate Character Detection","License plate character detection is a key step in intelligent transportation systems, demanding both high accuracy and low computational overhead for real-time edge deployment. Many leading deep-learning models achieve strong results using large architectures, which limits use on resource-constrained devices. This paper presents MicroCharNet, an ultra-lightweight end-to-end detector featuring a C2f-based backbone with CoordAtt, a C3k2 neck, and an anchor-free single-level head. Experiments on UFPR-ALPR show competitive accuracy with only 0.08M parameters and 0.096 GFLOPs, including hardware efficiency verification for real-time use.","MicroCharNet  \nLess is More for License Plate Character Detection  \nHuy Che 1, 2 , Dinh-Duy Phan 1, 2 and Duc-Lung Vu 1, 2,*  \n1University of Information Technology, Ho Chi Minh City, Vietnam  \n2Vietnam National University, Ho Chi Minh City, Vietnam  \n* Corresponding author: Duc-Lung Vu  \nEmail: [huycq@uit.edu.vn](huycq@uit.edu.vn), [duypd@uit.edu.vn](duypd@uit.edu.vn), [lungvd@uit.edu.vn](lungvd@uit.edu.vn)  \narXiv :2607 . 11830v1 [ cs .CV] 13 Jul 2026  \nAbstract—License plate character detection is a crucial component of intelligent transportation systems, where high accuracy and computational efficiency are required for realtime deployment. Although recent deep learning-based methods have substantially improved detection performance, many highaccuracy models rely on large-scale architectures that incur substantial computational overhead, limiting their applicability to resource-constrained devices. In this paper, we propose MicroCharNet, an ultra-lightweight model specifically designed for license plate character detection. The proposed architecture employs a compact backbone composed of C2f blocks, integrated with CoordAtt module to enhance feature extraction while preserving spatial information. A lightweight C3k2-based neck fuses multi-level features, followed by a single-level anchor-free detection head that enables end-to-end prediction. Experiments conducted on the UFPR-ALPR dataset demonstrate that MicroCharNet achieves competitive detection accuracy with only 0.08M parameters and 0.096 GFLOPs, while outperforming several recent YOLO-based baselines. Hardware-level evaluations further confirm its efficiency for real-time deployment on edge devices. These results indicate that carefully designed ultralightweight architectures can effectively balance accuracy and efficiency in license plate character detection. The source code is available at [https://github.com/chequanghuy/MicroCharNet](https://github.com/chequanghuy/MicroCharNet).  \nIndex Terms—Object Detection, License Plate, Computer Vision, YOLO, Embedded Device  \nI. INTRODUCTION  \nLicense plate recognition is a critical component in many intelligent transportation systems, including vehicle surveillance [1], smart parking [2], and traffic violation management [3] . In practical deployment scenarios, particularly on edge cameras or resource-constrained computing platforms, recognition models are required not only to achieve high accuracy but also to ensure fast inference for stable real-time operation.  \nIn recent years, deep learning models have significantly improved the performance of character and character-sequence recognition in natural images. However, most high-accuracy architectures are associated with a large number of parameters and high computational cost, making their direct deployment on resource-limited edge devices challenging. Among existing approaches, the YOLO family [4]–[9] provides an effective solution for real-time license plate character detection, thanks to its fast, flexible object detection. Nevertheless, since these architectures were originally designed for general-purpose computer vision tasks such as object detection [5], [6], [10], image segmentation [11], and multitask estimation [12], [13],  \nmAP@50  \nFig. 1. Accuracy–efficiency comparison between MicroCharNet and YOLObased baselines on the test set of UFPR-ALPR dataset.  \nthey are not truly optimized for the specific characteristics of license plate character detection. When applied directly, they often fail to fully exploit the fine-grained geometric features of characters, their sequential relationships, and the narrow regions of interest that are central to this task. As a result, lightweight networks trained on general-purpose vision tasks often do not achieve optimal performance when transferred to specialized application domains, such as license plate character detection [1]–[3], [10] .  \nUnlike approaches that prioritize large-scale models to maximize accuracy, this study ","cbCaij6jgOHNU2XJ","https://ap.wps.com/l/cbCaij6jgOHNU2XJ","pdf",1495925,4,1,6,"English","en",105,"# Introduction\n# Related Work\n# Proposed Method\n# Experiments\n# Conclusion","[{\"question\":\"What problem does MicroCharNet address?\",\"answer\":\"MicroCharNet targets license plate character detection in intelligent transportation systems, focusing on achieving accurate recognition with efficient computation suitable for real-time deployment.\"},{\"question\":\"What are the main architectural components of MicroCharNet?\",\"answer\":\"MicroCharNet uses a compact C2f-based backbone, integrates a CoordAtt module for enhanced feature extraction, adds a C3k2-based neck to fuse multi-level features, and applies a single-level anchor-free detection head for end-to-end prediction.\"},{\"question\":\"How does MicroCharNet perform in terms of efficiency and accuracy?\",\"answer\":\"On the UFPR-ALPR dataset, MicroCharNet achieves competitive detection accuracy while using only about 0.08M parameters and 0.096 GFLOPs, outperforming multiple recent YOLO-based baselines; hardware-level tests further support its real-time edge deployment capability.\"}]",1784210533,15,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"microcharnet-less-is-more-for-license-plate-character-detection","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"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/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/microcharnet-less-is-more-for-license-plate-character-detection/86333/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-26","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does MicroCharNet address?","Question",{"text":75,"@type":76},"MicroCharNet targets license plate character detection in intelligent transportation systems, focusing on achieving accurate recognition with efficient computation suitable for real-time deployment.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What are the main architectural components of MicroCharNet?",{"text":80,"@type":76},"MicroCharNet uses a compact C2f-based backbone, integrates a CoordAtt module for enhanced feature extraction, adds a C3k2-based neck to fuse multi-level features, and applies a single-level anchor-free detection head for end-to-end prediction.",{"name":82,"@type":73,"acceptedAnswer":83},"How does MicroCharNet perform in terms of efficiency and accuracy?",{"text":84,"@type":76},"On the UFPR-ALPR dataset, MicroCharNet achieves competitive detection accuracy while using only about 0.08M parameters and 0.096 GFLOPs, outperforming multiple recent YOLO-based baselines; hardware-level tests further support its real-time edge deployment capability.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,114,119,122,127,130,134],{"id":21,"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":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},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":106,"slug":137},19,"General","general"]