服装设计知识图谱中的服装装饰工艺分类模型

doi:10.13475/j.fzxb.20191003306

纺织学报 ›› 2020, Vol. 41 ›› Issue (08): 95-100.doi: 10.13475/j.fzxb.20191003306

服装设计知识图谱中的服装装饰工艺分类模型

杨娟¹^,², 张远鹏³^,⁴()

1.苏州大学纺织与服装工程学院, 江苏苏州 215123
2.南通大学纺织服装学院, 江苏南通 226001
3.南通大学智能信息技术研究中心, 江苏南通 226001
4.香港理工大学电子计算学系, 香港 999077

收稿日期:2019-10-16 修回日期:2020-04-27 出版日期:2020-08-15 发布日期:2020-08-21
通讯作者: 张远鹏
作者简介:杨娟(1983—),女,讲师,博士生。主要研究方向为服装知识图谱。
基金资助:
国家自然科学基金资助项目(81701793);江苏省哲学社会科学基金资助项目(18YSC009)

Garment ornamenting craft classification model for knowledge graph on clothing design

YANG Juan¹^,², ZHANG Yuanpeng³^,⁴()

1. College of Textile and Clothing Engineering, Soochow University, Suzhou, Jiangsu 215123, China
2. School of Textile and Clothing, Nantong University, Nantong, Jiangsu 226001, China
3. Research Institute of Smart Information Technology, Nantong University, Nantong, Jiangsu 226001, China
4. Departing of Computing, The Hong Kong Polytechnic University, Hong Kong 999077, China

Received:2019-10-16 Revised:2020-04-27 Online:2020-08-15 Published:2020-08-21
Contact: ZHANG Yuanpeng

摘要/Abstract

摘要：

为解决服装装饰工艺类别判定中噪声视角或弱相关视角带来的负影响问题,采用具备视角约减功能的多视角分类模型对噪声视角或弱相关视角进行动态过滤。首先基于1-阶TSK模糊系统,在其目标优化函数中引入视角间分类误差一致性约束,实现多视角协同学习;然后通过变体信息熵学习各视角的权重,并在权重学习过程中制定视角约减规则,自动剔除噪声视角或弱相关视角;最后通过服装装饰工艺类别分类实验对所构建模型的分类精度进行验证。结果表明:相比视角约减之前,所提出的多视角分类模型的测试精度提高了2.68%,可有效地过滤噪声视角或弱相关视角,降低其对分类精度的影响。

关键词: 服装装饰工艺, 模糊系统, 多视角数据, 视角约减, 协同学习, 服装设计资源, 知识图谱

Abstract:

In order to eliminate the negative influences caused by noisy views or weakly relevant views in the garment ornamenting craft classification tasks, an automatic view-reduction multi-view classification model was used to filter noisy views or weakly relevant views in this research. Based on the 1-order TSK fuzzy system, an error constraint item was introduced to be the objective function for collaborative learning. Then, a variant entropy item was designed to learn the weight of each view, and a reduction principle was designed to filter noisy views or weakly relevant views during collaborative learning. The proposed model was tested as the final step on the clothing ornamenting craft classification tasks. Experimental results demonstrate that the proposed classification model can reduce noisy views or weakly relevant views effectively such that the negative influences generated by them can be avoided. Compared with the model without view-reduction, the proposed classification model achieves a 2.68% improvement in terms of data accuracy.

Key words: garment ornamenting craft, fuzzy system, multi-view data, view-reduction, collaborative learning, clothing design resource, knowledge graph

中图分类号:

TP311.11

杨娟, 张远鹏. 服装设计知识图谱中的服装装饰工艺分类模型[J]. 纺织学报, 2020, 41(08): 95-100.

YANG Juan, ZHANG Yuanpeng. Garment ornamenting craft classification model for knowledge graph on clothing design[J]. Journal of Textile Research, 2020, 41(08): 95-100.

图/表 5

表1

图1

表2

表3

表4

对比分类模型的测试精度"

对比分类模型	视角约减前测试精度 (平均值±标准差)	约减视角数平均值	视角约减后测试精度 (平均值±标准差)
SVM	0.831 3±0.002 4	—	0.837 9±0.002 1
C4.5	0.809 7±0.010 5	—	0.811 1±0.009 8
1-TSK-FS	0.841 0±0.001 5	—	0.848 6±0.001 7
MV-TSK-FS	0.853 7±0.007 6	—	0.867 1±0.002 9
VR-MV-CM ( $δ m = 0$ )	0.852 1±0.003 4	—	0.877 8±0.002 0
VR-MV-CM ( $δ m ≠ 0$ )	0.878 9±0.001 6	11.4	—

表4

参考文献 13

[1]	我国古代印染工艺的发展史 [EB/OL]. (2011-03-31)[2011-03-31]. https://www.douban.com/group/topic/18723587/.
	The development history of printing and dyeing technology in ancient China[EB/OL]. (2011-03-31)[2011-03-31]. https://www.douban.com/group/topic/18723587/.
[2]	周捷, 马秋瑞. 基于BP神经网络的运动文胸肩带属性与乳房振幅的函数关系[J]. 纺织学报, 2019,40(9):186-191.
	ZHOU Jie, MA Qiurui. Prediction of relationship between shoulder strap attribute and breast amplitude of sports bra by BP neural network[J]. Journal of Textile Research, 2019,40(9):186-191.
[3]	王雯雯, 高畅, 刘基宏. 应用卷积神经网络的细纱断纱锭位识别[J]. 纺织学报, 2018,39(6):136-141.
	WANG Wenwen, GAO Chang, LIU Jihong. Position recognition of spinning yarn breakage based on convolution neural network[J]. Journal of Textile Research, 2018,39(6):136-141.
[4]	JIANG Yizhang, DENG Zhaohong, CHUNG Fulai, et al. Realizing two-view TSK fuzzy classification system by using collaborative learning[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2017,47(1):145-160. doi: 10.1109/TSMC.2016.2577558
[5]	张远鹏, 周洁, 邓赵红, 等. 代表点一致性约束的多视角模糊聚类算法[J]. 软件学报, 2019,30(2):282-301.
	ZHANG Yuanpeng, ZHOU Jie, DENG Zhaohong, et al. Multi-view fuzzy clustering approach based on medoid invariant constraint[J]. Journal of Software, 2019,30(2):282-301.
[6]	DENG Z H, JIANG Y Z, CHOI K, et al. Knowledge-leverage-based TSK fuzzy system modeling[J]. IEEE Transactions on Neural Networks and Learning Systems, 2013,24(8):1200-1212. doi: 10.1109/TNNLS.2013.2253617
[7]	杨娟, 吴志明, 张远鹏. 网络家用纺织品资源抽取方法[J]. 纺织学报, 2018,39(10):156-161.
	YANG Juan, WU Zhiming, ZHANG Yuanpeng. Extracting method of household textile resources from Web[J]. Journal of Textile Research, 2018,39(10):156-161.
[8]	王健, 张俊妮. 统计模型在中文文本挖掘中的应用[J]. 数理统计与管理, 2017,36(4):609-619.
	WANG Jian, ZHANG Junni. Applications of statistical models in Chinese text mining[J]. Journal of Applied Statistics and Management, 2017,36(4):609-619.
[9]	YAHAV I, SHEHORY O, SCHWARTZ D. Comments mining with TF-IDF: the inherent bias and its removal[J]. IEEE Transactions on Knowledge and Data Engineering, 2019,31(3):437-450. doi: 10.1109/TKDE.2018.2840127
[10]	JI S, SATISH N, LI S, et al. Parallelizing word2vec in shared and distributed memory[J]. IEEE Transactions on Parallel and Distributed Systems, 2019,30(9):2090-2100. doi: 10.1109/TPDS.71
[11]	WU X, ZUO W, LIN L, et al. F-SVM: combination of feature transformation and SVM learning via convex relaxation[J]. IEEE Transactions on Neural Networks and Learning Systems, 2018,29(11):5185-5199. doi: 10.1109/TNNLS.2018.2791507 pmid: 29994427
[12]	王文霞. 数据挖掘中改进的C4.5决策树分类算法[J]. 吉林大学学报(理学版), 2017,55(5):1274-1277.
	WANG Wenxia. Improved C4.5 decision tree classification algorithm in data mining[J]. Journal of Jilin University (Science Edition), 2017,55(5):1274-1277.
[13]	JIANG Yizhang, DENG Zhaohong, WU Dongrui, et al. Seizure classification from EEG signals using transfer learning, semi-supervised learning and tsk fuzzy system[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2017,25(12):2270-2284. doi: 10.1109/TNSRE.2017.2748388 pmid: 28880184

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

站名	站点网址
WGSN	https://www.wgsnchina.cn/cs/
POP服装趋势	https://www.pop-fashion.com
WOW TREND	http://www.wow-trend.com/
穿针引线	http://www.eeff.net/
CFW服装设计	https://art.cfw.cn/
YOKA时尚网	http://www.yoka.com/dna/
vogue	http://www.vogue.com.cn/
女装品牌网	http://lady.ef43.com.cn/
中国童装品牌网	http://www.61kids.com.cn/
肉丁网	https://www.rouding.com
华衣网	http://www.ef360.com/
编织人生	http://www.bianzhirensheng.com/
T100趋势	http://n.t100.cn/Product/index.aspx
蝶讯网	http://www.diexun.com/
印染在线	http://yr.e-dyer.com/
布流行	http://www.8liuxing.com/xinshou/
中国刺绣网	http://www.rocky168.com/
中国服饰新闻网倾城旗袍	http://www.cfw.com.cn/articlecat/id/79 https://www.hercity.com/
……	……

类别	文本
刺绣	刺绣工艺从刺绣方式上,可以分为3类:手工刺绣、缝纫机绣和电子控制机绣。这3种方法中,由于手工刺绣针法灵活多变,装饰效果好,适宜在各类服饰及织物上绣制……
印染	在中原地区,印花技术的再度复兴是从缬开始的,缬有绞缬、葛缬和夹缬。绞缬、葛缬实际上就是一种仿染印花的织物……
编结	编结,就是用线、绳编织出各种花样的网袋或饰物,古代俗称“绛子”或“络子”。我国古代的竹、柳、草、棕、藤编艺术更是源远流长,品种多样……
手绘	中国上古时期已在车舆、衣冠上绘画,作为某种标识。后来借鉴刺绣衣裳的美化效果,将细致的绣花版样描画在服装上,或将名人画稿描摹在服装上,成为手绘装饰服装……
镶嵌	服装制作工艺中的“镶拼”,指将2块或2块以上的布片连缀成一片,主要是指不同颜色、质地或不同纹理的布片在衣身、衣缝沿边等处以块面状的形式拼接……

对比分类模型	视角约减前训练精度 (平均值±标准差)	约减视角数平均值	视角约减后训练精度 (平均值±标准差)
SVM	0.894 1±0.001 4	—	0.899 7±0.003 6
C4.5	0.863 4±0.009 8	—	0.874 5±0.007 4
1-TSK-FS	0.865 0±0.003 2	—	0.869 8±0.004 5
MV-TSK-FS	0.898 7±0.008 4	—	0.914 5±0.002 7
VR-MV-CM (δ_m=0)	0.902 1±0.001 7	—	0.925 4±0.002 2
VR-MV-CM (δ_m≠0)	0.945 3±0.002 0	12.1	—

服装设计知识图谱中的服装装饰工艺分类模型

Garment ornamenting craft classification model for knowledge graph on clothing design

RichHTML

PDF (PC)

摘要/Abstract

引用本文

使用本文

图/表 5

参考文献 13

相关文章 0

Metrics

本文评价

推荐阅读 0