摘要
在零件生产的场景中往往需要对零件产品进行质量检测,多孔零件中的圆孔是否符合生产标准也是质量检测中的重要环节。为解决传统圆检测算法无法处理复杂场景下特定圆的检测问题,该文设计了一种基于卷积神经网络的零件圆检测方法,将圆检测分为3个阶段,第1阶段使用YOLOv5目标检测模型对零件图片中的目标圆进行粗检测,将多圆检测问题简化为单圆检测,获得含有单个目标圆的裁剪图片;第2阶段使用BiSeNet语义分割模型对单圆图片进行细检测,获得圆轮廓掩膜图;第3阶段使用改进的随机霍夫变换对圆参数进行检测,最终得到图中所有目标圆的半径与圆心坐标。经实验结果对比,该方法在多种阈值条件下的检测精度都高于其他对比方法,在IoU阈值为0.9的情况下F-meausre达到96%,能满足生产场景中的实时检测需求。
In the scene of parts production,it`s necessary to conduct quality inspection of parts products.Whether the round holes in porous parts meet the production standards is an important procedure in quality inspection.In order to solve the problem that traditional circle detection algorithms can hardly handle the detection of specific circles in complex scenes,we design a circle detection method based on convolutional neural network.The circle detection is divided into three stages.The first stage uses the YOLOv5 object detection model to perform rough detection on the target circle in the parts image,simplify the multi-circle detection problem into single-circle detection,and obtain a cropped image containing a single target circle.The second stage uses the BiSeNet semantic segmentation model to perform fine detection on the single-circle image,and obtains the circle contour binary map.The third stage uses improved randomized Hough transform to detect the circle parameters,and finally obtains the radius and center coordinates of target circles in the images.After comparing the experimental results,the detection accuracy of the proposed method is higher than that of other comparison methods under various threshold conditions,when IoU threshold is 0.9,F-meausre reaches 96%,which can meet the real-time detection requirements in production scenes.
作者
曾碧卿
杨睿
李一娴
张雅蓉
ZENG Bi-qing;YANG Rui;LI Yi-xian;ZHANG Ya-rong(School of Software,South China Normal University,Foshan 528225,China;Jihua Laboratory,Foshan 528200,China)
出处
《计算机技术与发展》
2023年第11期64-71,共8页
Computer Technology and Development
基金
国家自然科学基金面上项目(62076103)
广东省基础与应用基础研究基金项目(2021A1515011171)
广东省普通高校人工智能重点领域专项(2019KZDZX1033)。
关键词
圆检测
卷积神经网络
目标检测
语义分割
霍夫变换
circle detection
convolutional neural network
object detection
semantic segmentation
Hough transform
