[{"data":1,"prerenderedAt":21},["ShallowReactive",2],{"cms-blog-ai-vision-inspection-guide":3},{"id":4,"slug":5,"title":6,"title_en":7,"excerpt":8,"excerpt_en":9,"content":10,"category":11,"tags":12,"author":15,"cover_image":16,"published_at":17,"meta_title":16,"meta_description":16,"meta_keywords":16,"og_image":16,"status":18,"sort_order":19,"created_at":20,"updated_at":20},3,"ai-vision-inspection-guide","AI视觉检测系统实现方案","AI Visual Inspection System Implementation","工业质检场景下AI视觉检测的落地路径：数据采集、模型训练、边缘部署、效果评估。","AI visual inspection for industrial QC: data, training, edge deployment, evaluation.","\u003Cp>AI视觉检测正在替代传统人工质检。本文介绍从数据标注、模型选型、训练调优到边缘部署的完整流程。\u003C/p>\u003Ch3>数据准备\u003C/h3>\u003Cp>缺陷样本采集、标注规范、数据增强策略。\u003C/p>\u003Ch3>模型选型\u003C/h3>\u003Cp>YOLO、Faster R-CNN 等在工业场景的适用性对比。\u003C/p>\u003Ch3>边缘部署\u003C/h3>\u003Cp>模型量化、TensorRT 加速、产线实时推理。\u003C/p>","ai",[11,13,14],"视觉检测","工业","逸付技术",null,"2024-12-10 00:00:00","published",0,"2026-03-14T12:45:52.000Z",1776243288281]