机电工程技术2025，Vol.54Issue(6)：174-178,5.DOI:10.3969/j.issn.1009-9492.2025.06.030
基于3D视觉的EVA材质鞋面打孔关键技术研究
Key Technologies for EVA Shoe Punching Based on 3D Vision
摘要
Abstract
In recent years,some orders for EVA garden shoes have shifted to domestic production.Due to special process requirements,additional punching operations are needed on the curved surfaces with existing holes,which are currently completed manually,making automation impossible.To address this,3D vision technology and deep learning techniques have been introduced to study the key technologies for punching garden shoes.By integrating 3D vision acquisition technology,a 3D vision scanning system was constructed to obtain 3D point cloud data and grayscale information of the shoe surface.A deep learning model was then used to identify holes on the shoe surface,providing crucial information for subsequent punching operations.Furthermore,3D point cloud registration technology was applied to determine the positional relationship between the reference hole point cloud and the target hole point cloud,thereby obtaining the machinable poses for each hole.Finally,robotic technology was employed to automatically generate the punching programs for the garden shoe surfaces.This approach addresses the challenges posed by the deformable nature of garden shoe materials,which hinder automated processing,leading to high labor costs and labor shortages.It also accelerates the industry's progress toward automation and flexibility.关键词
三维成像/匹配算法/AI图像识别Key words
3D imaging/registration algorithm/AI image recognitiona分类
信息技术与安全科学引用本文复制引用
陈达伟..基于3D视觉的EVA材质鞋面打孔关键技术研究[J].机电工程技术,2025,54(6):174-178,5.基金项目
泉州市科技计划项目(2022C010R) （2022C010R）
