[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-464149-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-464149-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","an-automatic-detection-model-for-low-contrast-discrete-defects-on-aluminum-alloy-wheels-research-framework-and-results","An Automatic Detection Model for Low-Contrast Discrete Defects on Aluminum Alloy Wheels - Research Framework and Results","","X-ray-based non-destructive testing technology is essential for quality monitoring of aluminum alloy wheel hubs, yet casting images often show low contrast and discrete, edge-like defect distributions. Existing approaches struggle with limited feature extraction, inefficient cross-scale fusion, and sensitivity to complex background interference. This study introduces a complete detection framework that preprocesses images, uses an asymmetric pinwheel-shaped convolution (PConv) with an efficient receptive field for edge focus, and builds a Mamba-based two-stage feature pyramid network (MFDPN) for improved localization. A channel- and spatial-attention block (CASAB) further enhances robustness in complex scenes. On a self-built dataset, the method outperforms the baseline by 7.2% in mAP50 and 5% in Recall at 39 FPS, demonstrating practical value for automated defect detection.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/an-automatic-detection-model-for-low-contrast-discrete-defects-on-aluminum-alloy-wheels-research-framework-and-results/464149/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/an-automatic-detection-model-for-low-contrast-discrete-defects-on-aluminum-alloy-wheels-research-framework-and-results/464149.png","ImageObject",300,407,{"name":42,"@type":43},"LangkahRina","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-06","2026-09-30",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",9,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why are low-contrast discrete defects difficult to detect on aluminum alloy wheel hubs?","Question",{"text":63,"@type":64},"Wheel hub images from casting often have low contrast and defect edges are discretely distributed. Complex backgrounds can also interfere with feature extraction and localization.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What core modules does the proposed framework use?",{"text":68,"@type":64},"The framework includes data preprocessing, an asymmetric pinwheel-shaped convolution (PConv) for edge-focused features, a Mamba-based two-stage feature pyramid network (MFDPN) with a secondary focusing-diffusion mechanism, and a channel-and-spatial attention block (CASAB) to resist background interference.",{"name":70,"@type":61,"acceptedAnswer":71},"How does the proposed model perform on the self-built dataset?",{"text":72,"@type":64},"It achieves better performance than the baseline by 7.2% in mAP50 and 5% in Recall while running at 39 FPS inference speed, indicating good real-world usability.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},464149,1790795055,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":117,"language":139,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":12,"update_tm":143,"read_time":104,"preview_page_count":33,"preview_urls":144},962090893776,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","Article  \nAn Automatic Detection Model for Low-Contrast Discrete Defects on Aluminum Alloy Wheels  \nJian Yang , Ping Chen and Mingquan Wang *  \nAcademic Editor: Jiawei Xiang  \nReceived: 1 December 2025  \nRevised: 23 December 2025  \nAccepted: 25 December 2025  \nPublished: 26 December 2025  \nCopyright: © 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.  \nSchool of Information and Communication Engineering, North University of China, Taiyuan 030051, China; [b20240521@st.nuc.edu.cn](b20240521@st.nuc.edu.cn) (J.Y.); [b202505003@st.nuc.edu.cn](b202505003@st.nuc.edu.cn) (P.C.)  \n* [Correspondence: wangmq@nuc.edu.cn](Correspondence: wangmq@nuc.edu.cn)  \nAbstract  \nX-ray-based non-destructive testing technology plays a crucial role in the quality monitoring of aluminum alloy wheel hubs. Due to the characteristics of the casting process, wheel hub images often exhibit low contrast and a discrete distribution of defect edges. Existing methods often face problems such as poor feature extraction capability, low efficiency of cross-scale information fusion, and susceptibility to interference from complex backgrounds when detecting such defects. Therefore, this study proposes an innovative detection framework for defects in aluminum alloy wheel hubs. The model employs data preprocessing to enhance the quality of original images; integrates an asymmetric pinwheel-shaped convolution (PConv) with an efficient receptive field, enabling efficient focus on the edge feature information of discrete defects; innovatively constructs a Mamba-based two-stage feature pyramid network (MFDPN), which improves the network’s defect localization capability in complex scenarios via a secondary focusing-diffusion mechanism; and incorporates a channel and spatial attention block (CASAB), strengthening the model’s ability to resist interference from complex backgrounds. On our self-built wheel hub defect dataset, the proposed model outperforms the baseline by 7.2% in mAP50 and 5% in Recall at 39 FPS inference speed, thus validating its high practical utility for automated aluminum alloy wheel hub defect detection.  \nKeywords: defect detection; transformer; mamba; aluminum alloy wheels  \n1. Introduction  \nAs a crucial load-bearing and transmission component of automobiles, the quality of the wheel hub directly affects the safety and service life of the entire vehicle. In recent years, with the continuous improvement of requirements for automotive lightweighting and energy conservation and emission reduction, aluminum alloy—with its low density, excellent thermal and electrical conductivity, and good plasticity—has become the mainstream raw material for industrial wheel hub production [1] . The manufacturing processes of aluminum alloy wheels include low-pressure casting, gravity casting, forging and spinning forming, etc. Low-pressure casting has higher production efficiency, better dimensional consistency and lower defect rate of finished products compared with other methods, and has become one of the main processes for manufacturing aluminum alloy wheels [2] . However, due to the complexity and inherent characteristics of the casting process itself, aluminum alloy wheels are highly prone to defects during the die-casting process [3] . These defects not only reduce the mechanical properties of the wheels but may also cause safety  \nhazards during use [4] . Early detection of substandard products can effectively prevent safety accidents and reduce maintenance costs for enterprises in the later stage.  \nNon-destructive testing (NDT) refers to the inspection methods that identify surface and internal defects of materials or components without damaging the inspected object or affecting its subsequent performance. At present, commonly used non-destructive testing methods include ultrasonic testing [5], infrared thermal imaging [6","cbCaidv3CNPdje2K","https://ap.wps.com/l/cbCaidv3CNPdje2K","pdf",67059833,"English","# Abstract\n# 1. Introduction\n## Background and importance of wheel hub quality\n## Non-destructive testing methods and X-ray advantages\n## Manual inspection limitations\n## Deep learning approaches for aluminum alloy defect detection\n## Related work and evaluation criteria","[{\"question\":\"Why are low-contrast discrete defects difficult to detect on aluminum alloy wheel hubs?\",\"answer\":\"Wheel hub images from casting often have low contrast and defect edges are discretely distributed. Complex backgrounds can also interfere with feature extraction and localization.\"},{\"question\":\"What core modules does the proposed framework use?\",\"answer\":\"The framework includes data preprocessing, an asymmetric pinwheel-shaped convolution (PConv) for edge-focused features, a Mamba-based two-stage feature pyramid network (MFDPN) with a secondary focusing-diffusion mechanism, and a channel-and-spatial attention block (CASAB) to resist background interference.\"},{\"question\":\"How does the proposed model perform on the self-built dataset?\",\"answer\":\"It achieves better performance than the baseline by 7.2% in mAP50 and 5% in Recall while running at 39 FPS inference speed, indicating good real-world usability.\"}]","An Automatic Detection Model for Low-Contrast Discrete Defects on Aluminum Alloy Wheels - Research Framework and Results | PDF",1790767748,[145,149,153,157],{"page_no":22,"url":146,"width":147,"height":148},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/abacdfd2279e66f8b06c0ab40ea1e3fc.webp",983,1390,{"page_no":26,"url":150,"width":151,"height":152},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/23a7160ab499bfde8b6a4d91c442566c.webp",982,1388,{"page_no":30,"url":154,"width":155,"height":156},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/e3f563b946196d3e4fd614defd0e64aa.webp",981,1387,{"page_no":33,"url":158,"width":159,"height":160},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/5b566e3edc9f24b4738b38483b4f4084.webp",980,1385]