计算机应用 ›› 2019, Vol. 39 ›› Issue (9): 2580-2585.DOI: 10.11772/j.issn.1001-9081.2019030523

• 人工智能 • 上一篇    下一篇

结合全卷积网络和K均值聚类的球栅阵列焊球边缘气泡分割

赵瑞祥1,2, 侯宏花1,2, 张鹏程1,2, 刘祎1,2, 田珠1,2, 桂志国1,2   

  1. 1. 中北大学 信息与通信工程学院, 太原 030051;
    2. 生物医学成像与影像大数据山西省重点实验室(中北大学), 太原 030051
  • 收稿日期:2019-03-29 修回日期:2019-05-18 发布日期:2019-05-31 出版日期:2019-09-10
  • 通讯作者: 桂志国
  • 作者简介:赵瑞祥(1994-),男,山西临汾人,硕士研究生,主要研究方向:图像处理、深度学习;侯宏花(1975-),女,副教授,博士,主要研究方向:医学图像处理、计算机视觉;张鹏程(1984-),男,内蒙古巴彦淖尔人,讲师,博士,主要研究方向:精准放射治疗剂量计算及方案优化;刘祎(1987-),女,河南睢县人,副教授,博士,主要研究方向:图像处理、医学图像重建;田珠(1995-),男,山西长治人,硕士研究生,主要研究方向:图像分割、深度学习;桂志国(1972-),男,天津蓟县人,教授,博士,主要研究方向:信号与信息处理、图像处理和识别、图像重建。
  • 基金资助:

    国家重大科学仪器设备开发专项(2014YQ24044508);国家自然科学基金资助项目(61671413,61801438);中北大学青年学术带头人项目(QX201801);山西省应用基础研究项目(201801D221196)。

Welding ball edge bubble segmentation for ball grid array based on full convolutional network and K-means clustering

ZHAO Ruixiang<sup>1,2</sup>, HOU Honghua<sup>1,2</sup>, ZHANG Pengcheng<sup>1,2</sup>, LIU Yi<sup>1,2</sup>, TIAN Zhu<sup>1,2</sup>, GUI Zhiguo<sup>1,2</sup>   

  1. 1. School of Information and Communication Engineering, North University of China, Taiyuan Shanxi 030051, China;
    2. Shanxi Provincial Key Laboratory for Biomedical Imaging and Big Data(North University of China), Taiyuan Shanxi 030051, China
  • Received:2019-03-29 Revised:2019-05-18 Online:2019-05-31 Published:2019-09-10
  • Supported by:

    This work is partially supported by the National Major Scientific Instruments and Equipment Development Special (2014YQ24044508); the National Natural Science Foundation of China (61671413, 61801438); the Middle School Senior Academic Leadership Program (QX201801); the Shanxi Applied Basic Research Project (201801D221196).

摘要:

针对在球栅阵列(BGA)气泡检测中,由于图像干扰因素的多样性导致焊球存在边缘气泡与背景之间灰度级接近,从而造成焊球气泡分割结果不精确的问题,提出了一种结合全卷积神经网络(FCN)和K均值(K-means)聚类分割的焊球气泡分割方法。首先根据所制作的BGA标签数据集搭建FCN,通过训练该网络得到合适的网络模型,再对待测BGA图像进行预测处理得到图像的粗分割结果;然后对焊球区域映射提取,通过同态滤波法提高气泡区域辨识度,再使用K-means聚类分割对图像进行细分割处理,得到最终分割结果图;最后对原图焊球及气泡区域进行标注识别。将所提出的算法与传统BGA气泡分割算法进行对比,实验结果表明,所提出的算法对复杂BGA焊球的边缘气泡分割精确,图像分割结果与其真实轮廓高度匹配,准确度更高。

关键词: 全卷积网络, K均值聚类, 球栅阵列, 边缘气泡, 缺陷分割

Abstract:

For inaccurate segmentation results caused by the existence of edge bubbles in welding balls and the grayscale approximation of background due to the diversity of image interference factors in Ball Grid Array (BGA) bubble detection, a welding ball bubble segmentation method based on Fully Convolutional Network (FCN) and K-means clustering was proposed. Firstly, a FCN network was constructed based on the BGA label dataset, and trained to obtain an appropriate network model, and then the rough segmentation result of the image were obtained by predicting and processing the BGA image to be detected. Secondly, the welding ball region mapping was extracted, the bubble region identification was improved by homomorphic filtering method, and then the image was subdivided by K-means clustering segmentation to obtain the final segmentation result. Finally, the welding balls and bubble region in the original image were labeled and identified. Comparing the proposed algorithm with the traditional BGA bubble segmentation algorithm, the experimental results show that the proposed algorithm can segment the edge bubbles of complex BGA welding balls accurately, and the image segmentation results highly match the true contour with higher accuracy.

Key words: Full Convolutional Network (FCN), K-means clustering, Ball Grid Array (BGA), edge bubble, defect segmentation

中图分类号: