The fiber producer needed to replace slow, physically demanding manual inspection of spun-yarn products with a more consistent and scalable quality process, because long hours of flashlight-based visual checking were hurting quality, staffing, and energy efficiency.
Reported impact
The fiber plant reduced manual inspection effort, accelerated quality checks, improved consistency, and created a data foundation that can support wider intelligent-manufacturing upgrades.
Baidu AI Cloud helped Hengyi Petrochemical's Shuangtu New Material operation replace manual fiber inspection with an AI-driven machine-vision workflow. Along the line, industrial cameras capture multiple images from each bobbin and send them to a data center, where models analyze defects such as hair yarn, knots, oil stains, and other surface problems. The system then classifies suspicious products and feeds sorting decisions back to the production line so only true exceptions require human review. To make the models work in practice, experienced quality staff were pulled into the loop as data annotators, converting long-held defect-recognition knowledge into reusable training data. The result is not just faster visual inspection; it is a new feedback loop in which production images, labeled defect data, and continuous model tuning create a repeatable quality-control asset that can support broader manufacturing modernization across the fiber operation.
Twenty industrial cameras capture images of bobbins on the line, split each image into many subregions, and send them to a central analysis system. Defect-classification models judge whether a bobbin is abnormal, route suspect products for review, and use human-validated labels to improve future detection performance and production decision-making.
AI can inspect a 20-kilogram bobbin in 2.5 seconds, cutting time by more than 70% (original: 20公斤丝饼的质检只需要2.5秒，耗时比传统人工检测缩短了70%以上).
The system identifies defects such as hair yarn, tangled yarn, spider-web yarn, oil stains, and knots (original: 后台利用人工智能模型算法可以对毛丝、绊丝、蛛网丝、油污、夹结丝等缺陷进行准确检测).
Twenty industrial cameras shoot the line and each bobbin generates more than ten photos for analysis (original: 20台工业相机拍摄流水线上的两锭丝锭，每个丝锭会生成10余张照片).
Workers became AI data annotators so the machine could learn product-defect recognition from expert experience (original: 王礼娜成为一名人工智能数据标注师…将质检员的工作经验转化成数据，让人工智能设备学会辨别产品缺陷).
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
Baidu AI Cloud Sample Apr 9, 2026 China Baowu Steel Group Baowu needed to move ladle status data, image data and production-process data safely to the cloud so it could detect hidden risks, optimize operations and lower storage costs. BML Full-Function AI Development Platform · Big Data Platform · IoT Core Baidu AI Cloud Sample Apr 8, 2026 SANY Heavy Machinery Co., Ltd. (三一重机) SANY needed to uphold its 'solve within one day' service standard for excavators and other machinery even when remote field engineers lacked deep experience and half of faults could not be diagnosed accurately without… Before the project, as many as 50% of problems could not be diagnosed accurately. Industrial AR Remote Assistance System · Intelligent Industrial Quality Inspection · Tiangong Edge-Cloud Integrated IoT Platform Baidu AI Cloud Sample Dec 26, 2025 China Iron and Steel Research Institute Group (中国钢研科技集团) China Iron and Steel Research needed a practical way to apply large-model and visual AI capabilities across metallurgical research and manufacturing, especially for metallography analysis, surface defect detection, and… 100 ms response time Baige AI Heterogeneous Computing Platform · Kunlun · Qianfan Large Model Platform 