基于图像光影重构的缎纹影光织物明度预测方法

doi:10.13475/j.fzxb.20210500207

纺织学报 ›› 2022, Vol. 43 ›› Issue (05): 97-103.doi: 10.13475/j.fzxb.20210500207

基于图像光影重构的缎纹影光织物明度预测方法

郑雯洁¹, 张爱丹¹^,²()

1.浙江理工大学纺织科学与工程学院(国际丝绸学院), 浙江杭州 310018
2.浙江理工大学浙江省丝绸与时尚文化研究中心, 浙江杭州 310018

收稿日期:2021-05-06 修回日期:2022-01-25 出版日期:2022-05-15 发布日期:2022-05-30
通讯作者: 张爱丹
作者简介:郑雯洁(1996—),女,硕士生。主要研究方向为织物图像处理与纺织品设计。
基金资助:
浙江省丝绸与时尚文化研究中心培育项目(ZSFCRC202208PY)

Lightness prediction method for shaded satin fabrics based on image reconstruction of light and shadow

ZHENG Wenjie¹, ZHANG Aidan¹^,²()

1. College of Textile Science and Engineering(International Institute of Silk), Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China
2. Silk and Fashion Culture Research Center of Zhejiang Province, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China

Received:2021-05-06 Revised:2022-01-25 Published:2022-05-15 Online:2022-05-30
Contact: ZHANG Aidan

摘要/Abstract

摘要：

为更准确获取缎纹影光组织织物的明度值,提出一种基于织物图像光影重构的明度预测方法。将织物图像分离为经纬图形层、阴影层和材质层,计算3层中不同因素的相对面积率,并赋予各图形面积以实际经纬纱线明度值,采用明度相加重构织物图像明度值,再将重构明度与织物实际测色数据进行拟合分析。选择经纬图形、投影和杂点面积率为自变量,建立经、纬两向加强织物的明度预测回归模型,并随机选取样本检验模型精确度。结果表明:经纬图形层重构明度与织物实测明度的接近度为0.15,依次增加阴影层和材质层明度,则接近度从0.76提升为0.89;模型预测的织物明度总体相对误差在±4%以内,预测结果可为影光组织织物设计提供参考。

关键词: 缎纹影光组织, 图像重构, 回归分析, 织物明度, 预测模型

Abstract:

In order to obtain the brightness value of shaded satin fabrics more accurately, a lightness prediction method based on the shadow reconstruction of fabric image was proposed. The fabric image was separated into three layers: warp and weft graphic layer, shadow layer and texture layer, and the actual color values of the warp and the weft yarn were assigned to the relative area of the three separated image layers when their pixel numbers were counted respectively. After the fabric image brightness value were reconstructed by adding the lightness values, fitting analysis were implemented between the reconstructed one and the actual color measurement data of the fabric. According to the analysis results, the warp and weft area rate, shadow area rate and texture area rate were selected as the independent variables. A regression model for the lightness prediction of the fabric was established based on the three independent variables, and random samples were selected to test the accuracy of the model. The results show that the proximity between the lightness of the warp and weft graphic layer and the measured lightness is 0.15, and the proximity increases from 0.76 to 0.89 when the shadow layer and material layer are added in sequence. The overall relative error of the fabric lightness prediction by the model is within ±4%, the prediction results can be used as references for the design of the fabric with shaded weaves.

Key words: shaded satin weave, image reconstruction, regression analysis, fabric brightness, prediction model

中图分类号:

TS141.9

郑雯洁, 张爱丹. 基于图像光影重构的缎纹影光织物明度预测方法[J]. 纺织学报, 2022, 43(05): 97-103.

ZHENG Wenjie, ZHANG Aidan. Lightness prediction method for shaded satin fabrics based on image reconstruction of light and shadow[J]. Journal of Textile Research, 2022, 43(05): 97-103.

图/表 6

图1

图2

图3

表1

表2

表3

参考文献 12

[1]	傅艺扬, 刘妹琴, 樊臻, 等. 基于纹理滤波和颜色聚类的提花织物纹样自动提取方法[J]. 丝绸, 2019, 12(56): 9-15.
	FU Yiyang, LIU Meiqin, FAN Zhen, et al. Automatic pattern extraction of jacquard fabric based on texture filtering and color clustering[J]. Journal of Silk, 2019, 12(56): 9-15.
[2]	化春键, 孙康康, 陈莹. 基于改进FCM聚类的网孔织物图像分割算法[J]. 光电子·激光, 2020, 31(8): 857-864.
	HUA Chunjian, SUN Kangkang, CHEN Ying. Method of mesh fabric image segmentation based on improved fuzzy C-means clustering[J]. Journal of Optoelectro-nics·Laser, 2020, 31(8): 857-864.
[3]	赵艳, 左保齐. 机器视觉在织物疵点检测上的应用研究综述[J]. 计算机工程与应用, 2020, 56(2): 11-17.
	ZHAO Yan, ZUO Baoqi. Analysis on application of machine vision in fabric fefect detection[J]. Computer Engineering and Applications, 2020, 56(2): 11-17.
[4]	周赳, 罗秉芬, 叶莹洁. 以影光组织为基础的高花效果提花织物设计[J]. 纺织学报, 2017, 38(5): 49-52.
	ZHOU Jiu, LUO Bingfen, YE Yingjie. Design of high pattern jacquard fabric based on shaded weaves[J]. Journal of Textile Research, 2017, 38(5): 49-52.
[5]	周赳, 白琳琳. 纬二重渐变全遮盖结构设计研究与实践[J]. 纺织学报, 2018, 39(1): 32-37.
	ZHOU Jiu, BAI Linlin. Design research and practice on gradient weft-full-backed structure[J]. Journal of Textile Research, 2018, 39(1): 32-37.
[6]	罗来丽, 胡丁亭, 唐澜倩, 等. 基于全显色结构的三组纬提花织物混色特征研究[J]. 丝绸, 2011, 48(10): 28-32.
	LUO Laili, HU Dingting, TANG Lanqian, et al. Research on color mixture characteristics of triple-weft jacquard fabric with full-color structure[J]. Journal of Silk, 2011, 48(10): 28-32.
[7]	黄紫娟. 一经一纬单层色织物明度模型建立初探[J]. 国外丝绸, 2008(4): 33-34,40.
	HUANG Zijuan. A preliminary study on the establish-ment of lightness model of single layer color fabric at one weft[J]. Silk Textile Technology Overseas, 2008(4): 33-34,40.
[8]	罗来丽, 王春燕, 周赳. 基于全显色结构的二组纬提花织物的混色特征[J]. 纺织学报, 2012, 33(4): 39-44.
	LUO Laili, WANG Chunyan, ZHOU Jiu. Research on mixed color characteristics of double weft jacquard fabric with colored wefts[J]. Journal of Textile Research, 2012, 33(4): 39-44.
[9]	杨丹, 赵海滨, 龙哲. MatLab图像处理实例详解[M]. 北京: 清华大学出版社, 2013:345.
	YANG Dan, ZHAO Haibin, LONG Zhe. Detailed illustration of image processing in MatLab[M]. Beijing: Tsinghua University Press, 2013:345.
[10]	周赳, 吴文正. 无彩数码提花织物的创新设计原理和方法[J]. 纺织学报, 2006, 29(4): 1-5.
	ZHOU Jiu, FRANKIE Ng. Innovative principle and method of colorless digital jacquard fabric design[J]. Journal of Textile Research, 2006, 29(4): 1-5. doi: 10.1177/004051755902900101
[11]	姚金波, 王志意, 董敏, 等. 温湿环境下棉织物的抗皱性能变化规律及其预测模型[J]. 天津工业大学学报, 2020, 39(5): 43-49.
	YAO Jinbo, WANG Zhiyi, DONG Min, et al. Change and prediction model of crease resistance of cotton fabric under warm and humid environment[J]. Journal of Tiangong University, 2020, 39(5): 43-49.
[12]	杨阳, 俞欣, 章为敬, 等. 针织面料凉爽性能的评价方法及其预测模型[J]. 纺织学报, 2021, 42 (3): 95-101.
	YANG Yang, YU Xin, ZHANG Weijing, et al. Evaluation method and prediction model establishment of cooling performance of knitted fabrics[J]. Journal of Textile Research, 2021, 42 (3): 95-101.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

试样编号	L₁与L_pi		L₁与L_ti		L₁与L_zi
试样编号	F值	p值	F值	p值	F值	p值
1^*	2.53	0.123	0.06	0.812	0.00	0.948
2^*	1.51	0.229	0.27	0.608	0.06	0.815
3^*	3.21	0.084	0.03	0.871	0.01	0.917

模型编号	R²	调整R²	F值	剔除变量
1^#	0.985	0.982	336.008	x₃、x₄
2^#	0.961	0.958	343.01	x₁、x₄、x₇
3^#	0.953	0.950	344.864	x₁、x₃、x₄、x₇

样品编号	x₁	x₂	x₃	实测数据	预测数据	误差率%
1	2.382	0.299	0.035	50.84	49.82	-2.01
2	2.994	0.409	0.012	65.37	64.59	-0.57
3	2.330	0.219	0.023	52.55	52.08	-0.89
4	2.522	0.377	0.006	68.15	68.80	0.96
5	2.319	0.204	0.039	44.92	44.56	-0.81
6	3.096	0.408	0.006	71.02	69.93	-1.93
7	2.225	0.173	0.009	58.71	59.27	0.96
8	5.834	0.468	0.005	73.91	72.76	-1.55
9	2.255	0.162	0.034	34.97	36.14	3.36
10	2.276	0.256	0.010	60.49	59.86	-1.05
11	2.268	0.260	0.048	38.36	37.59	-2.01
12	2.410	0.210	0.008	63.55	63.20	-0.56

[1]	王利君, 马希明, 丁殷佳, 陈诚毅. 风速对单双层着装状态下运动服针织面料湿阻的影响[J]. 纺织学报, 2021, 42(07): 151-157.
[2]	杨阳, 俞欣, 章为敬, 张佩华. 针织面料凉爽性能的评价方法及其预测模型[J]. 纺织学报, 2021, 42(03): 95-101.
[3]	张晓侠, 刘凤坤, 买巍, 马崇启. 基于BP神经网络及其改进算法的织机效率预测[J]. 纺织学报, 2020, 41(08): 121-127.
[4]	黄倩倩, 李俊. 环境温度突变时人体热感觉变化机制研究进展[J]. 纺织学报, 2020, 41(04): 188-194.
[5]	刘红, 丁雪梅, 吴雄英. 滚筒洗涤负载类型对织物运动及其洗涤性能的影响[J]. 纺织学报, 2019, 40(06): 44-49.
[6]	周捷, 马秋瑞. BP神经网络在塑身内衣压力预测中的应用[J]. 纺织学报, 2019, 40(04): 111-116.
[7]	王利君毛鹏丽. 防电磁辐射聚吡咯/ 棉织物的制备及其性能[J]. 纺织学报, 2018, 39(09): 95-101.
[8]	吴莹汪军. 基于K-SVD学习字典的机织物纹理表征及应用[J]. 纺织学报, 2018, 39(02): 165-170.
[9]	罗璇蒋高明丛洪莲. 横编毛衫原料消耗预测模型[J]. 纺织学报, 2017, 38(02): 159-164.
[10]	高融王玲杨以雄. 服装线上零售渠道消费者决策机制[J]. 纺织学报, 2016, 37(12): 146-152.
[11]	周小溪梁惠娥陈潇潇董稚雅秦潇璇. 春夏季衬衫用色织面料材质的感觉评价[J]. 纺织学报, 2016, 37(08): 59-64.
[12]	陈丽华. 织物定伸长力与弹力服装宽裕量的关系[J]. 纺织学报, 2016, 37(07): 115-0120.
[13]	陶建勤余晓华张宏伟刘娜. 毛纱捻缩率的预测模型[J]. 纺织学报, 2016, 37(05): 32-36.
[14]	王晓明邹婷李超婧王璐. 基于编织点起始位置及牵拉速度变化的编织角预测模型[J]. 纺织学报, 2015, 36(09): 28-33.
[15]	丛洪莲徐巧金兰名. 咖啡炭/聚酯长丝纬编提花床垫面料的性能[J]. 纺织学报, 2015, 36(03): 37-42.

基于图像光影重构的缎纹影光织物明度预测方法

Lightness prediction method for shaded satin fabrics based on image reconstruction of light and shadow

RichHTML

PDF (PC)

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 12

相关文章 15

Metrics

本文评价

推荐阅读 0