纺织学报 ›› 2019, Vol. 40 ›› Issue (07): 174-181.doi: 10.13475/j.fzxb.20180900808

• 管理与信息化 • 上一篇    下一篇

结合显著区域检测和手绘草图的服装图像检索

吴传彬1, 刘骊1,2(), 付晓东1,2, 刘利军1,2, 黄青松1,2   

  1. 1. 昆明理工大学 信息工程与自动化学院, 云南 昆明 650500
    2. 昆明理工大学 云南省计算机技术应用重点实验室, 云南 昆明 650500
  • 收稿日期:2018-09-03 修回日期:2019-04-15 出版日期:2019-07-15 发布日期:2019-07-25
  • 通讯作者: 刘骊
  • 作者简介:吴传彬(1993-),男,硕士生。主要研究方向为计算机视觉、图像处理。
  • 基金资助:
    国家自然科学基金项目(61862036);国家自然科学基金项目(61462051);国家自然科学基金项目(61462056);国家自然科学基金项目(81560296);云南省应用研究基础计划面上项目(2017FB097)

Clothing image retrieval by salient region detection and sketches

WU Chuanbin1, LIU Li1,2(), FU Xiaodong1,2, LIU Lijun1,2, HUANG Qingsong1,2   

  1. 1. Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, Yunnan 650500, China
    2. Computer Technology Application Key Laboratory of Yunnan Province, Kunming University of Science and Technology, Kunming, Yunnan 650500, China
  • Received:2018-09-03 Revised:2019-04-15 Online:2019-07-15 Published:2019-07-25
  • Contact: LIU Li

RichHTML

9

PDF (PC)

82

摘要：

针对服装图像检索准确率和效率较低的问题,提出一种服装显著区域检测和手绘草图的服装图像检索方法。首先采用正则化随机漫步算法对输入的服装图像库进行视觉显著区域检测,并结合其边缘轮廓信息,得到服装显著边缘图像;其次,对输入的服装草图和服装边缘图像进行特征提取,得到服装草图和服装边缘图像各自的方向梯度直方图(HOG)特征;然后,通过计算服装草图特征和服装边缘特征的相似度,实现特征匹配;最后,按照特征匹配结果在服装图像库中检索与服装草图相似的服装图像,采用基于距离相关系数的重排序算法对其相似度进行排序并输出检索结果。结果表明,该方法提高了服装检索的准确率,具有较好的鲁棒性,检索准确率可达78.5%。

关键词: 服装检索, 手绘草图的图像检索, 显著性检测, 特征匹配

Abstract:

In order to solve the problems of unsatisfactory accuracy and low efficiency in the clothing image retrieval, a sketch based clothing image retrieval method by visual salient regions and re-ranking was proposed. Firstly, clothing salient edge map was obtained by saliency detection method with regularized random walks walking and the edge map. Then, histogram of oriented gradeient features of user sketches and the salient edge in clothing images were extracted, respectively, and the feature matching was achieved by similarity calculation between the input sketches and clothing images. Finally, the retrieval results were output in descending order according to the similarity. Using the re-ranking optimization based on distance correlation coefficients, final results were obtained. Experimental results show that the method can effectively provide clothing retrieval results and significantly improve accuracy and robustness comparison with other approaches. The accuracy ratio of the algorithm is higher than 78.5%.

Key words: clothing retrieval, sketch-based image retrieval, saliency detection, feature matching

中图分类号: 

  • TP391.41

图1

本文方法流程图"

图2

GF-HOG特征提取图"

图3

草图数据"

表1

检测结果对比"

方法 召回率 精确率 F值
MR 0.852 0.658 0.745
RC 0.596 0.921 0.728
本文方法 0.813 0.821 0.826

图4

显著区域检测结果"

图5

本文方法与其他方法的比较"

图6

字典大小的影响"

图7

参数w大小的影响"

图8

没有使用重排序的检索结果"

图9

使用重排序后的检索结果"

表2

几种算法的NDCG(K20)比较"

方法 NDCG(K20) 时间/s
文献[9] 0.658 3.18
文献[10] 0.391 5.56
HOG[13] 0.367 6.53
Edgel[18] 0.280 2.96
HLR[19] 0.584 3.23
RST-SHELO[20] 0.679 2.39
本文方法 0.785 2.42

图10

手绘连体裤在不同风格下的检索结果"

[1] KIAPOUR M H, HAN X, LAZEBNIK S, et al. Where to buy it: matching street clothing photos in online shops[C]// IEEE International Conference on Computer Vision. Piscataway: IEEE Xplore, 2015: 3343-3351.
[2] LIU Z, LUO P, QIU S, et al. DeepFashion: powering robust clothes recognition and retrieval with rich annotations[C]// Computer Vision and Pattern Recognition. Piscataway: IEEE Xplore, 2016: 1096-1104.
[3] TSENG C H, HUNG S S, TSAY J J . An efficient garment visual search based on shape context[J]. Wseas Transactions on Computers, 2009,8(7):223-230.
[4] KONDO S, TOYOURA M, MAO X, et al. Sketch based skirt image retrieval[C]// Non Photorealistic Animation and Rendering. New York: ACM Digital Library, 2014: 11-16.
[5] 李东, 万贤福, 汪军 . 采用傅里叶描述子和支持向量机的服装款式识别方法[J]. 纺织学报, 2017,38(5):122-127.
LI Dong, WAN Xianfu, WANG Jun . Clothing style recongnition approach using Fourier descriptor and support vector machines[J]. Journal of Textile Research, 2017,38(5):122-127.
[6] 项明, 姚雪存, 崔振东 , 等. 利用相关反馈技术和形状特征的服装图像检索算法[J]. 纺织学报, 2014,35(2):94-98.
XIANG Ming, YAO Xuecun, CUI Zhendong , et al. Research on image retrieval algorithm by relevance feedback technology and shape features[J]. Journal of Textile Research, 2014,35(2):94-0.
[7] LI K, PANG K, SONG Y Z, et al. Fine-grained sketch-based image retrieval: the role of part-aware attri-butes[C]// Applications of Computer Vision. Piscataway: IEEE Xplore, 2016: 1-9.
[8] AZIZPOUR H, LAPTEV I. Object detection using strongly-supervised deformable part models[C]// European Conference on Computer Vision. Berlin: Springer-Verlag, 2014: 836-849.
[9] ZHANG Y, QIAN X, TAN X , et al. Sketch-based image retrieval by salient contour reinforcement[J]. IEEE Transactions on Multimedia, 2016,18(8):1604-1615.
[10] PEDRONETTE D C G, TORRES R D S . Exploiting clustering approaches for image re-ranking[J]. Journal of Visual Languages & Computing, 2011,22(6):453-466.
[11] ARBELÁEZ P, MAIRE M, FOWLKES C , et al. Contour detection and hierarchical image segment-ation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2011,33(5):898-916.
[12] ACHANTA R, SHAJI A, SMITH K , et al. SLIC superpixels compared to state-of-the-art superpixel methods.[J]. IEEE Trans Pattern Anal Mach Intell, 2012,34(11):2274-2282.
doi: 10.1109/TPAMI.2012.120 pmid: 22641706
[13] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]// IEEE Computer Society. Piscataway: IEEE Xplore, 2005: 886-893.
[14] HU R, COLLOMOSSE J . A performance evaluation of gradient field HOG descriptor for sketch based image retrieval[J]. Computer Vision & Image Understanding Cviu, 2013,117(7):790-806.
[15] SZÉKELY G J, RIZZO M L, BAKIROV N K. Measuring and testing dependence by correlation of distances.[C]// ACM Symposium on Virtual Reality Software and Technology. New York: ACM Digital Library, 2007: 141-144.
[16] YANG C, ZHANG L, LU H, et al. Saliency detection via graph-based manifold ranking[C]// IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society. Piscataway: IEEE Xplore, 2013: 3166-3173.
[17] CHENG M M, ZHANG G X, MITRA N J, et al. Global contrast based salient region detection[C]// Computer Vision and Pattern Recognition. Piscataway: IEEE Xplore, 2011: 409-416.
[18] CAO Y, WANG C, ZHANG L, et al. Edgel index for large-scale sketch-based image search[C]// IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Xplore, 2011: 761-768.
[19] WANG S, ZHANG J, HAN T X , et al. Sketch-based image retrieval through hypojournal-driven object boundary selection with HLR descriptor[J]. IEEE Transactions on Multimedia, 2015,17(7):1045-1057.
[20] SAAVEDRA J M , RST-SHELO. Sketch-based image retrieval using sketch tokens and square root normalization[J]. Multimedia Tools & Applications, 2017,76(1):931-951.
[21] SEZGIN T M, STAHOVICH T, DAVIS R. Sketch based interfaces:early processing for sketch under-standing[C]// The Workshop on Perceptive User Interfaces. New York: ACM Digital Library, 2001: 1-8.
[1] 汪会, 孙洁, 丁笑君, 龙颖, 邹奉元. 织物纹样特征提取与匹配方法比较[J]. 纺织学报, 2020, 41(04): 45-50.
[2] 王文胜, 李天剑, 冉宇辰, 卢影, 黄民. 筒子纱纱笼纱杆的定位检测方法[J]. 纺织学报, 2020, 41(03): 160-167.
[3] 曹霞 李岳阳 罗海驰 蒋高明 丛洪莲. 蕾丝花边的改进型纹理特征检索方法[J]. 纺织学报, 2016, 37(06): 142-154.
[4] 康锋 胡洁 张华熊 周慧. 基于最大稳定极值区域的织物图案检索[J]. 纺织学报, 2015, 36(10): 156-160.
[5] 林翔宇 张华熊. 基于图像边缘特征的领带花型检索方法[J]. 纺织学报, 2015, 36(05): 123-126.
[6] 刘军;李重;胡觉亮. 基于特征匹配的三维试衣效果[J]. 纺织学报, 2009, 30(01): 122-126.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
京ICP备14006648号　 京公网安备11010502044800号
版权所有 © 2021 《纺织学报》编辑部
地址： 北京市朝阳区延静里中街3号主楼6层《纺织学报》杂志社　邮政编码：100025　
电话：010-65017711,65917740　传真：010-65016539　Email:fangzhixuebao@vip.126.com
本系统由北京玛格泰克科技发展有限公司设计开发