基于机器视觉的零件自动识别研究

详细信息    查看官网全文

• 英文论文题名：A Study of Automatic Component Recognition Algorithm Based on Machine Vision
• 论文作者：张超 ; 李慧
• 英文论文作者：Zhang Chao ; Li Hui ; China Aviation Planning and Design Institute(Group) CO.LTD.
• 年：2016
• 作者机构：中国航空规划设计研究总院有限公司技术院;
• 论文关键词：零件识别 ; 特征提取 ; SURF ; 主成分分析
• 英文论文关键词：component recognition ; feature extraction ; SURF ; principal component analysis
• 会议召开时间：2016-10-31
• 会议录名称：探索 创新 交流（第7集）——第七届中国航空学会青年科技论坛文集（下册）
• 英文会议录名称：Discovery,Innovation and Communication——Proceedings of 7th CSAA Science and Technology Youth Forum
• 语种：中文
• 分类号：TP391.41
• 学会代码：ZGHU
• 会议名称：探索 创新 交流（第7集）——第七届中国航空学会青年科技论坛
• 会议地点：中国广东中山
• 主办单位：中国航空学会
• 学会名称：中国航空学会
• 页数：5
• 文件大小：2121k
• 原文格式：O
• 会议级别：全国

摘要

为快速、准确地进行零件图像识别,提出一种基于SURF的特征识别算法。为了提高识别率,先对零件图像进行预处理,包括图像增强,中值滤波。然后由SURF算法获取零件图像的特征点和相应的特征向量。鉴于特征向量的高维特性,对特征向量用主成分分析降维后采用最近邻距离比率双向匹配算法,并采用随机抽样一致算法进一步提纯匹配点,从而实现零件识别。实验结果表明,本文算法可以有效应对零件图片在发生旋转变化、尺度变化、模糊变化和亮度变化后的识别问题,具有较好的鲁棒性和实时性,能够满足工业化要求。
To achieve fast and precise component recognition,an automatic target recognition method is proposed based on speed up robust feature algorithm.To increase recognition rale,the component image is pre-processed via image enhancement and median filter.The feature points of the component image and corresponding eigenvectors are obtained by SURF algorithm.Due to the high dimensional eigenvectors,feature descriptors are established and principle component analysis is employed to reduce the dimensionality.Finally,nearest neighbor distance ratio classifier is employed in dual-directions and mis-matches are eliminated by random sample consensus.The experiment results demonstrates that the proposed algorithm could solve the recognition problems caused by rotation,scale,blur and illumination transformation and also exhibit promising robustness and real-time,which can meet the requirements in real-time performance and for industrial application.

引文

[1]毕文波.基于机器视觉的机械零件尺寸识别研究[D].山东:山东大学,2006.
    [2]张辉,王耀南,周博文,葛继.医药大输液可见异物自动视觉检测方法及系统研究[J].电子测量与仪器学报,2010,24(2):125-130.
    [3]Lowe D G.Distinctive Image Feature from Scale Invariant Keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
    [4]魏振山,赵长宽.基于傅里叶特征的几何零件快速识别[J].机械设计与制造,2010(6):54-56.
    [5]晏开华,苏真伟.支持向量机在机械零件识别中的应用[J].计算机技术与应用,2008(11):108-110,114.
    [6]Bay H,Tuytelaars T,Gool L V.Surf:Speed-up Robust Features[C].Proceedings of the 9th European Conference on Computer Vision.2006:404-417.
    [7]李弼程,鹏天强,彭波等.智能图像处理技术[M].北京:电子工业出版社,2004.
    [8]Bay H,Ess A,Tuytelaars T,Van Gool L.Speeded-up Robust Features(surf)[J].Comput Vis Image Underst,2008,110(3):346-359.
    [9]Viola P,Jones M J.Rapid Object Detection Using a Boosted Cascade of Simple Features[C]//Conference on Computer Vision and Pattern Recognition.2001;511-518.
    [10]Lindeberg T.Scale-Space theory in computer vision[M].Springer,1993.
    [11]Strang G.Wavelet Transforms Versus Fourier Transforms[J].Bulletin of the American Mathematical Society,1993,28(2):288-305.
    [12]Strang G.ffavelet Transforms Versus Fourier Transforms[J].Bulletin of the American Mathematical Society,1993,28(2):288-305.