基于改进全卷积神经网络的图像单像素边缘提取

doi:10.19678/j.issn.1000-3428.0053574

计算机工程 ›› 2020, Vol. 46 ›› Issue (1): 262-270. doi: 10.19678/j.issn.1000-3428.0053574

基于改进全卷积神经网络的图像单像素边缘提取

刘畅, 张剑, 林建平

同济大学机械与能源工程学院, 上海 201804

收稿日期:2019-01-04 修回日期:2019-02-25 出版日期:2020-01-15 发布日期:2019-03-05
作者简介:刘畅(1995-),男,硕士研究生,主研方向为计算机视觉;张剑(通信作者),副教授、博士;林建平,教授、博士。
基金资助:
2017年智能制造综合标准化与新模式应用项目"磁性材料智能制造新模式应用"。

Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network

LIU Chang, ZHANG Jian, LIN Jianping

School of Mechanical Engineering, Tongji University, Shanghai 201804, China

Received:2019-01-04 Revised:2019-02-25 Online:2020-01-15 Published:2019-03-05

摘要/Abstract

摘要： 为实现复杂背景图像中高精度边缘的准确提取，提出一种改进的单像素边缘提取算法。在改进的全卷积神经网络中，通过添加辅助输出层与采取多尺度输入的方式初步提取图像多像素边缘，并利用分水岭算法对多像素边缘进行细化重定位，从而获取图像单像素边缘。磁瓦图像上的应用结果表明，该算法具有较强的鲁棒性，能提取高精度且完整连续的单像素边缘。

关键词: 单像素边缘检测, 全卷积神经网络, 分水岭算法, 距离误差, 磁瓦图像

Abstract: To accurately extract high precision edges in complex background images,this paper proposes an improved single-pixel edge extraction algorithm.In the improved fully convolutional neural network,this method adds an auxiliary output layer and adopts a multi-scale input method to coarsely extract multi-pixel edges of an image.Then the watershed algorithm is used to refine and relocate the multi-pixel edges to obtain a high precision single-pixel edge of an image.Application results on magnetic tile images show that the algorithm has strong robustness and can extract complete continuous high precision single-pixel edges.

Key words: single-pixel edge detection, fully convolutional neural network, watershed algorithm, distance error, magnetic tile image

中图分类号:

TP391.4

刘畅, 张剑, 林建平. 基于改进全卷积神经网络的图像单像素边缘提取[J]. 计算机工程, 2020, 46(1): 262-270.

LIU Chang, ZHANG Jian, LIN Jianping. Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network[J]. Computer Engineering, 2020, 46(1): 262-270.

http://www.ecice06.com/CN/Y2020/V46/I1/262

图/表 14

20200115111521

20200115111523

20200115111526

20200115111528

20200115111531

20200115111533

20200115111536

20200115111539

20200115111542

20200115111551

20200115111601

20200115111604

20200115111608

20200115111611

参考文献

[1] JIA Yonghong.Digital image processing[M].2nd ed.Wuhan:Wuhan University Press.2010.(in Chinese)贾永红.数字图像处理[M].2版.武汉:武汉大学出版社,2010.
[2] JIANG Honghai,LI Xueqin,LIU Peiyong,et al.Detection method of typical defects in arc ferrite magnet surface[J].Journal of Southwest Jiaotong University,2013,48(1):129-134.(in Chinese)蒋红海,李雪琴,刘培勇,等.铁氧体磁瓦表面典型缺陷检测方法[J].西南交通大学学报,2013,48(1):129-134.
[3] CANNY J.A computational approach to edge detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,8(6):679-698.
[4] SHI Cong.Magnets surface defect inspection method base on computer vision[D].Harbin:Harbin Institute of Technology,2013.(in Chinese)石聪.磁材表面缺陷视觉检测方法[D].哈尔滨:哈尔滨工业大学,2013.
[5] DUDA R,HART P,STROK D.Pattern classification and scene analysis[J].IEEE Transactions on Automatic Control,2003,19(4):462-463.
[6] MU Kenan,ZHAO Xiangmo,HUI Fei.Multiscale fused edge detection algorithm based on non-sampling difference of Gaussian pyramid[J].Journal of Sichuan University(Engineering Science Edition),2015,47(5):130-138.(in Chinese)穆柯楠,赵祥模,惠飞.基于非采样高斯差分金字塔的多尺度融合边缘检测[J].四川大学学报(工程科学版),2015,47(5):130-138.
[7] LI Zheng,ZHANG Hai.An adaptive edge detection approach based on the gradient direction consistency[J].Journal of Electronics and Information Technology,2016,38(7):1674-1681.(in Chinese)李正,张海.一种基于梯度方向一致性的边缘检测新算法[J].电子与信息学报,2016,38(7):1674-1681.
[8] WANG Fuping,SHUI Penglang.Edge detection algorithm using shape-adaptive anisotropic defferential filter[J].Systems Engineering and Electronics,2016,38(12):2876-2883.(in Chinese)王富平,水鹏朗.形状自适应各向异性微分滤波器边缘检测算法[J].系统工程与电子技术,2016,38(12):2876-2883.
[9] ZHANG Wenkun,WANG Xiyuan,SONG Jiaqian.Fractional differential difference and Gaussian curvature filter edge detection algorithm[J].Computer Engineering,2019,45(2):213-219.(in Chinese)张文坤,汪西原,宋佳乾.基于分数阶微分差与高斯曲率滤波的边缘检测算法[J].计算机工程,2019,45(2):213-219.
[10] GANIN Y,LEMPITSKY V.N⁴-fields:fields:neural network nearest neighbor fields for image transforms[C]//Proceedings of Asian Conference on Computer Vision.Berlin,Germany:Springer,2014:536-551.
[11] BERTASIUS G,SHI J,TORRESANI L.DeepEdge:a multi-scale bifurcated deep network for top-down contour detection[C]//Proceeding of 2015 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2015:1-5.
[12] SHEN Wei,WANG Xinggang,WANG Yan,et al.DeepContour:a deep convolutional feature learned by positive-sharing loss for contour detection[C]//Proceeding of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2015:3982-3991.
[13] XIE Saining,TU Zhuowen.Holistically-nested edge detection[J].International Journal of Computer Vision,2015,125(1/2/3):3-18.
[14] LIU Y,LEW M S.Learning relaxed deep supervision for better edge detection[C]//Proceeding of Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:1-8.
[15] LIU Yun,CHENG Mingming,HU Xiaowei,et al.Richer convolutional features for edge detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2019,41(8):5872-5881.
[16] VINCENT L,SOILLE P.Watersheds in digital spaces:an efficient algorithm based on immersion simulations[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1991,13(6):583-598.
[17] SZEGEDY C,LIU N W,JIA N Y,et al.Going deeper with convolutions[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2015:1-8.
[18] OGNIEWICZ R L,KUBLER O.Hierarchic voronoi skeletons[J].Pattern Recognition,1995,28(3):343-359.
[19] AUGUST J,SIDDIQI K,ZUCKER S W.Ligature instabilities in the perceptual organization of shape[J].Computer Vision and Image Understanding,1999,76(3):231-243.
[20] YAN Tingqin,ZHOU Changxiong,LIU Shufen.A universal thinning algorithm restricted by distance[J].Journal of Nanjing University (Natural Sciences),2013,49(2):189-195.(in Chinese)颜廷秦,周昌雄,刘淑芬.一种距离场约束下的普适细化算法[J].南京大学学报(自然科学版),2013,49(2):189-195.
[21] BENJAMINI Y,HOCHBERG Y.Controlling the false discovery rate:a practical and powerful approach to multiple testing[J].Journal of the Royal Statistical Society,1995,57(1):289-300.

选择文件类型/文献管理软件名称

选择包含的内容

基于改进全卷积神经网络的图像单像素边缘提取

Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献

相关文章 14

编辑推荐

Metrics

本文评价

[1]	冯玉芳, 殷宏, 卢厚清, 程恺, 曹林, 刘满. 基于改进全卷积神经网络的红外与可见光图像融合方法[J]. 计算机工程, 2020, 46(8): 243-249,257.
[2]	梅旭璋, 江红, 孙军. 基于密集注意力网络的视网膜血管图像分割[J]. 计算机工程, 2020, 46(3): 267-272,279.
[3]	刘天宇, 姜威威, 何江萍, 韩金仓. 基于HC-CFCN模型的肝脏CT图像分割[J]. 计算机工程, 2020, 46(2): 268-273.
[4]	亢洁, 丁菊敏, 万永, 雷涛. 基于分水岭修正与U-Net的肝脏图像分割算法[J]. 计算机工程, 2020, 46(1): 255-261,270.
[5]	张杰妹, 杨词慧. 基于RV-FCN的CT肝脏影像自动分割算法[J]. 计算机工程, 2019, 45(7): 258-263.
[6]	肖锋,李茹娜. 语义信息引导下的显著目标检测算法[J]. 计算机工程, 2019, 45(4): 248-253.
[7]	张永梅,付昊天,孙海燕,张睿,陈立潮,潘理虎. 一种基于改进FCN的多光谱图像建筑物识别方法[J]. 计算机工程, 2019, 45(1): 239-245.
[8]	赵雅坤,徐娟,蒋娇龙. 基于RSSI与移动锚节点运动轨迹的定位算法[J]. 计算机工程, 2019, 45(1): 35-39.
[9]	黄应清, 齐鸥, 蒋晓瑜, 刘中晅. 基于分水岭算法的图像融合质量评价方法[J]. 计算机工程, 2013, 39(1): 252-255.
[10]	武宗茜, 王鹏, 丁天怀. 活动轮廓模型在重叠藻细胞计数中的应用[J]. 计算机工程, 2012, 38(3): 209-211.
[11]	陈百英, 刘以安, 张强. 改进UKF算法在天波超视距雷达中的应用[J]. 计算机工程, 2012, 38(24): 78-80.
[12]	邹志文, 张立强, 陈继明, 梁军. 结合网格分割和边折叠的网格简化算法[J]. 计算机工程, 2012, 38(17): 245-248.
[13]	宋建军, 侯志强, 余旺盛. 基于人类视觉特性的梯度骨架标记提取算法[J]. 计算机工程, 2012, 38(10): 9-13.
[14]	刁智华, 赵春江, 郭新宇, 陆声链, 王秀徽. 分水岭算法的改进方法研究[J]. 计算机工程, 2010, 36(17): 4-6.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于改进全卷积神经网络的图像单像素边缘提取

Single-Pixel Edge Extraction of Image Based on Improved Fully Convolutional Neural Network

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献

相关文章 14

编辑推荐

Metrics

本文评价