基于OpenCV图像处理的拥抱模拟压力自适应服装设计

doi:10.13475/j.fzxb.20251006501

摘要/Abstract

摘要：

围绕触觉交互需求,针对现有拥抱模拟服装反馈模块定位模糊、压力输出模式固化及互动形式单一等问题,选取具有高情感表达强度的拥抱行为作为研究对象,对亲密关系人群的拥抱姿势展开系统研究,并集成柔性可穿戴设备,设计并开发了一款拥抱模拟压力自适应服装。选取手部位置与躯体姿态作为关键拥抱特征,采用可逆感温变色涂层面料实验服装,结合拍摄记录法采集拥抱图像,通过OpenCV图像处理获得正、背、双侧和肩部共18张成色平均图像,并结合主观数据生成的压力感知热力图,最终划分得到拥抱综合区域定位。在此基础上,以ESP32为核心控制器,电阻式压力传感器作为感应输入单元,气囊作为压力反馈执行单元,开发了具备自适应压力调节功能的拥抱模拟服装。经功能测试,该服装的压力自适应识别在实际预设、系统识别与个人感知三者间完全一致率达60%。本研究不仅为拥抱模拟服装中反馈模块的布局提供了方法与图像依据,也为触觉交互服装与情感可穿戴设备的产品化开发提供了技术参考与实践路径。

关键词: 情感交互, 模拟拥抱, 触觉反馈, OpenCV, 柔性气囊, 压力传感器, 功能服装, 嵌入式系统

Abstract:

Objective Hugging, with its broad emotional expression and strong psychological identification, is a common posture in haptic interaction. Various feedback methods, different interaction scenarios, and diverse devices have been used for wearable devices for simulating hugs. However, current devices, especially garment-type ones, have limitations. Feedback module placement lacks clear visual reference, pressure output struggles to adapt to different partners, and functional barriers exist between devices designed for collaborative use and those for individual use. These issues hinder the devices' universality and social interaction capabilities. Exploring hugging postures is important for optimizing feedback module placement and for improving device interaction methods.

Method This study focuses on images of hugging postures among individuals in intimate relationships. By integrating subjective evaluations with objective image capture, average images were analyzed to propose segmentation and localization of comprehensive hugging zones and develop pressure-adaptive garment for simulated hugs. Based on hand placement and body position, three common and comfortable combinations were selected as variables, which are the right lateralization criss-cross style, the neck-waist style with the neck surrounded, and neck-waist style with the waist surrounded. Objective image capture was facilitated utilizing experimental garments coated with reversible thermochromic paint, capturing front, back, sides, and shoulder region images by photographic documentation. Subjective evaluations were adopted to collect participants' pressure perceived zones for ranking the body parts during hugging through online questionnaires. The functional garment employed an ESP32 microcontroller as the main control unit, flexible resistive pressure sensors as input units, and flexible airbag as feedback actuators.

Results For the 360 images collected from objective experiments, multi-image averaging was performed using OpenCV, yielding a total of 18 average images, corresponding to the spatial distribution characteristics of the front, back, sides, and shoulder parts under three hugging postures. Based on the 18 average images, two comprehensive average images were further synthesized, presenting the overall average distribution of the front and back views for the three postures. Analysis of subjective data generated 18 pressure perception heatmaps, reflecting the distribution of pressure perception intensity on different body parts during the hugging process. SPSS was adopted to analyze the ranking data of body parts, and the results revealed a high degree of consistency in the perception intensity priorities between the front and back parts under different hugging postures. In contrast, the perception intensity of the shoulder region showed lower consistency. By comparing the average images with the pressure perception heatmaps, comprehensive hugging zones were manually delineated. Based on this, the spatial positioning and overall layout of the pressure feedback modules for functional garments were determined. Functional testing revealed that the pressure-adaptive recognition of this garment achieved a 60% consistency rate on three metrics, i.e., actual preset values, system recognition, and personal perception.

Conclusion This study focuses on hugging postures among individuals in intimate relationships, employing OpenCV average image processing and subjective pressure perception heatmaps to explore the spatial distribution characteristics of three hugging postures. Manual segmentation is adopted to delineate comprehensive hugging zones. Based on the extracted zones, a functional hugging simulation garment was designed and developed, which was capable of adaptive adjustment according to varying hugging pressure. The findings provide methodological and visual foundations for designing feedback modules in simulated hugging garments, establishing a reusable methodological framework for future development and optimization of functional hugging devices. Furthermore, adhering to the law of large numbers where increasing sample size drives sample means toward theoretical expectations, subsequent research can enhance the representativeness of average images and the universality of region localization by expanding image acquisition scale and establishing broader standardized image databases.

Key words: affective interaction, simulated hug, haptic feedback, OpenCV, flexible airbag, pressure sensor, functional garment, embedded system

中图分类号:

TS941.731

温轶涵, 鲁虹. 基于OpenCV图像处理的拥抱模拟压力自适应服装设计[J]. 纺织学报, 2026, 47(05): 190-200.

WEN Yihan, LU Hong. Design of pressure adaptive garment for simulated hugs based on OpenCV image processing[J]. Journal of Textile Research, 2026, 47(05): 190-200.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.fzxb.org.cn/CN/10.13475/j.fzxb.20251006501

http://www.fzxb.org.cn/CN/Y2026/V47/I05/190

图/表 14

图1

图2

图3

图4

表1

图5

图6

表2

图7

图8

图9

表3

表4

表5

参考文献 30

[1]	FORSELL L M, ÅSTRÖM J A. Meanings of hugging: from greeting behavior to touching implications[J]. Comprehensive Psychology, 2012, 1: 02.17.21.CP.1.13.
[2]	COHEN S, JANICKI-DEVERTS D, TURNER R B, et al. Does hugging provide stress-buffering social support: a study of susceptibility to upper respiratory infection and illness[J]. Psychological Science, 2015, 26(2): 135-147. doi: 10.1177/0956797614559284
[3]	VAN RAALTE L J, FLOYD K. Daily hugging predicts lower levels of two proinflammatory cytokines[J]. Western Journal of Communication, 2021, 85(4): 487-506. doi: 10.1080/10570314.2020.1850851
[4]	LIGHT K C, GREWEN K M, AMICO J A. More frequent partner hugs and higher oxytocin levels are linked to lower blood pressure and heart rate in premenopausal women[J]. Biological Psychology, 2005, 69(1): 5-21. pmid: 15740822
[5]	GREWEN K M, GIRDLER S S, AMICO J, et al. Effects of partner support on resting oxytocin, cortisol, norepinephrine, and blood pressure before and after warm partner contact[J]. Psychosomatic Medicine, 2005, 67(4): 531-538. pmid: 16046364
[6]	MALLOUK N, ROSHANFAR M, FEKRI P, et al. A brief review on recent advances in haptic technology for human-computer interaction: force, tactile, and surface haptic feedback[C]// 2024 7th International Conference on Information and Computer Technologies (ICICT). New York: IEEE, 2024: 34-39.
[7]	MORIKAWA O, HASHIMOTO S, MUNAKATA T, et al. Embrace system for remote counseling[C]// Proceedings of the 8th International Conference on Multimodal Interfaces. New York: ACM, 2006: 318-325.
[8]	ARAFSHA F, ALAM K M, EL SADDIK A. EmoJacket: consumer centric wearable affective jacket to enhance emotional immersion[C]// 2012 International Conference on Innovations in Information Technology (IIT). New York: IEEE, 2012: 350-355.
[9]	ISRAR A, POUPYREV I. Tactile brush: drawing on skin with a tactile grid display[C]// Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. New York: ACM, 2011: 2019-2028.
[10]	KONISHI Y, HANAMITSU N, MINAMIZAWA K, et al. Synesthesia suit: the full body immersive experience[C]// ACM SIGGRAPH 2016 Posters. New York: ACM, 2016: 1.
[11]	NOBUHIRO T, RYUTA O, HIROYUKI O, et al. Sense-roid: emotional haptic communication with yourself[C]// Proceedings of Virtual Reality International Conference 2011 (VRIC 2011). Laval, France; 2011: 1-4.
[12]	BROWN H, SHANNON E, FITTER N T. Design and preliminary evaluation of the AID vest: an automatic inflatable wearable device for anxiety reduction[C]// 2021 IEEE World Haptics Conference (WHC). New York: IEEE, 2021: 745-750.
[13]	杨乐丹, 刘运娟, 麻文效, 等. 基于接触安慰原理的模拟拥抱功能性服装设计与开发[J]. 丝绸, 2025, 62(1): 77-84.
	YANG Ledan, LIU Yunjuan, MA Wenxiao, et al. Design and development of simulated hugging functional clothing based on the principle of contact comfort[J]. Journal of Silk, 2025, 62(1): 77-84.
[14]	TEH J K S, TSAI Z, KOH J T K V, et al. Mobile implementation and user evaluation of the Huggy Pajama system[C]// 2012 IEEE Haptics Symposium (HAPTICS). New York: IEEE, 2012: 471-478.
[15]	TSETSERUKOU D. HaptiHug: a novel haptic display for communication of hug over a distance[C]// Haptics: Generating and Perceiving Tangible Sensations. Berlin, Heidelberg: Springer, 2010: 340-347.
[16]	CHAN A. ″I need a hug right now″: affective support through remote touch technology[C]// 2019 8th International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW). New York: IEEE, 2019: 40-44.
[17]	LIN W K, ZHANG D S, LEE W W, et al. Super-resolution wearable electrotactile rendering system[J]. Science Advances, 2022, 8(36): eabp8738.
[18]	LI J Y, CHEN T S, MA S Y, et al. PuppetAI: a customizable platform for designing tactile-rich affective robot interaction[EB/OL]. 2026: arXiv: 2602.04787. https://arxiv.org/abs/2602.04787.
[19]	DELAZIO A, NAKAGAKI K, KLATZKY R L, et al. Force jacket: pneumatically-actuated jacket for embodied haptic experiences[C]// Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems. New York: ACM, 2018: 1-12.
[20]	MUELLER F, VETERE F, GIBBS M R, et al. Hug over a distance[C]// CHI05 Extended Abstracts on Human Factors in Computing Systems. New York: ACM, 2005: 1673-1676.
[21]	JIANG M Q, NANJAPPAN V, LI Z M, et al. iPillowPal: exploring affective gesture-based interactive textile design for long-distance relationship couples[J]. International Journal of Human-Computer Studies, 2025, 199: 103466. doi: 10.1016/j.ijhcs.2025.103466
[22]	BLOCK A E, KUCHENBECKER K J. Softness, warmth, and responsiveness improve robot hugs[J]. International Journal of Social Robotics, 2019, 11(1): 49-64. doi: 10.1007/s12369-018-0495-2
[23]	DROLET M, CAMPBELL J, BEN AMOR H. Learning and blending robot hugging behaviors in time and space[C]// 2023 IEEE International Conference on Robotics and Automation (ICRA). New York: IEEE, 2023: 12071-12077.
[24]	DUEREN A L, VAFEIADOU A, EDGAR C, et al. The influence of duration, arm crossing style, gender, and emotional closeness on hugging behaviour[J]. Acta Psychologica, 2021, 221: 103441. doi: 10.1016/j.actpsy.2021.103441
[25]	FLOYD K. All touches are not created equal: effects of form and duration on observers' interpretations of an embrace[J]. Journal of Nonverbal Behavior, 1999, 23(4): 283-299. doi: 10.1023/A:1021602926270
[26]	TURNBULL O H, STEIN L, LUCAS M D. Lateral preferences in adult embracing: a test of the ″hemispheric asymmetry″ theory of infant cradling[J]. The Journal of Genetic Psychology, 1995, 156(1): 17-21. doi: 10.1080/00221325.1995.9914802
[27]	OCKLENBURG S, PACKHEISER J, SCHMITZ J, et al. Hugs and kisses: the role of motor preferences and emotional lateralization for hemispheric asymmetries in human social touch[J]. Neuroscience & Biobehavioral Reviews, 2018, 95: 353-360. doi: 10.1016/j.neubiorev.2018.10.007
[28]	PACKHEISER J, ROOK N, DURSUN Z, et al. Embracing your emotions: affective state impacts lateralisation of human embraces[J]. Psychological Research, 2019, 83(1): 26-36. doi: 10.1007/s00426-018-0985-8 pmid: 29349505
[29]	FAUL F, ERDFELDER E, LANG A G, et al. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences[J]. Behavior Research Methods, 2007, 39(2): 175-191. doi: 10.3758/bf03193146 pmid: 17695343
[30]	YANG Y, REN L, CHEN C, et al. SnapInflatables: designing inflatables with snap-through instability for responsive interaction[C]// Proceedings of the CHI Conference on Human Factors in Computing Systems. New York: ACM, 2024: 1-15.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

躯干部位	排序赋值 (标准化)	N(x)			卡方检验
躯干部位	排序赋值 (标准化)	右偏向十字交叉	颈部被环绕	腰部被环绕	卡方检验
正面	1	24(60.0)	23(57.5)	20(50.0)	ns
	2	7(17.5)	2(5.0)	4(10.0)
	3	5(12.5)	13(32.5)	12(30.0)
	4	4(10.0)	2(5.0)	4(10.0)
背面	1	11(27.5)	4(10.0)	13(32.5)	ns
	2	17(42.5)	18(45.0)	20(50.0)
	3	9(22.5)	15(37.5)	5(12.5)
	4	3(7.5)	3(7.5)	2(5.0)
肩部	1	4(10.0)	12(30.0)	1(2.5)	^***
	2	10(25.0)	16(40.0)	8(20.0)
	3	17(42.5)	8(20.0)	11(27.5)
	4	9(22.5)	4(10.0)	20(50.0)

电子元件名称	主要功能
ESP32-WROOM-32芯片	系统编译运算核心
Micro SD卡转接模块	网页代码读取通道
PCA9685舵机驱动板	I²C通信协议控制气泵模块
380微型气泵	实现充气与放气
DS3231时钟模块	获取精确时间
电阻式薄膜压力传感器	获取拥抱压力数据
多通道采集板	压力传感器集成通道

评估对象	评级
评估对象	实际预设	系统识别	个人感知
1^#	3	3	2
2^#	3	3	1
3^#	3	3	2
4^#	3	3	3
5^#	3	3	2
6^#	2	3	3
7^#	2	2	3
8^#	2	2	3
9^#	2	2	2
10^#	2	2	1
11^#	1	2	1
12^#	1	1	1
13^#	1	1	2
14^#	1	1	1
15^#	1	1	3

对象组合	Cohen's Kappa值		线性加权Kappa系数
对象组合	数值	解释	数值	解释
实际预设& 系统识别	0.800	高度一致	0.850	高度一致
系统识别& 个人感知	0.000	随机一致	0.100	轻微一致
实际预设& 个人感知	0.000	随机一致	0.100	轻微一致

对象序号	轻压力拥抱				中压力拥抱				强压力拥抱
对象序号	A	B	C	D	A	B	C	D	A	B	C	D
1	3	3	3	2	7	7	7	6	6	6	6	5
2	6	7	6	6	6	6	7	5	6	5	6	4
3	6	6	6	5	6	6	6	4	6	6	6	4
4	7	4	6	4	6	6	5	6	6	3	3	3
5	7	6	5	6	5	6	7	7	6	7	6	6
平均分	5.8	5.2	5.2	4.6	6.0	6.2	6.4	5.6	6.0	5.4	5.4	4.4