[1]耿增民,刘诗宇,康申奥,等.基于残差网络注意力机制的民族服饰图像分类[J].服装学报,2025,10(06):550-555.
　GENG Zengmin,LIU Shiyu,KANG Shenao,et al.Ethnic Costume Image Classification Based on Residual Network Attention Mechanism[J].Journal of Clothing Research,2025,10(06):550-555.
点击复制

基于残差网络注意力机制的民族服饰图像分类()

《服装学报》[ISSN:2096-1928/CN:32-1864/TS]

卷:
第10卷
期数:
2025年06期
页码:
550-555
栏目:
民族服装
出版日期:
2025-12-30

文章信息/Info

Title:
Ethnic Costume Image Classification Based on Residual Network Attention Mechanism
作者:
耿增民1;  刘诗宇2;  康申奥3;  高丹丹4;  崔 牮2;  钟 鸣3
1.北京服装学院 人工智能与创新设计学院,北京 100029; 2.北京服装学院 信息中心,北京 100029; 3. 北京服装学院 服装艺术与工程学院,北京 100029; 4.北京服装学院 民族服饰博物馆,北京 100029
Author(s):
GENG Zengmin1;  LIU Shiyu2;  KANG Shen’ao3;  GAO Dandan4;  CUI Jian2;  ZHONG Ming3
1.School of Artificial Intelligence and Innovative Design, Beijing Institute of Fashion Technology, Beijing 100029,China; 2.Information Center, Beijing Institute of Fashion Technology, Beijing 100029,China; 3. School of Fashion Art and Engineering, Beijing Institute of Fashion Technology, Beijing 100029,China; 4. Museum of Ethnic Costumes, Beijing Institute of Fashion Technology, Beijing 100029,China
分类号:
TP 183; TS 941.26
文献标志码:
A
摘要:
针对民族服饰图像因纹理复杂、特征多样而导致自动分类精度不足、难以满足当前民族服饰的数字化保护与传承需求的问题,采用融合卷积注意力机制的残差网络方法,构建一种改进的ResNet-50模型,以提升模型在民族服饰图像特征捕捉和泛化分类方面的能力。基于包含汉族、苗族、蒙古族、满族、藏族5类民族服饰共15 000张图像的数据集,将改进模型与ResNet-34、ResNet-50、VGG16、AlexNet等模型进行分类性能对比。结果显示,改进模型在准确率、精确率、召回率及F1分数等多项评价指标上均显著优于其他模型,提升了民族服饰图像的自动化分类能力,为民族服饰数字化保护与传承提供了一种高效、可靠的新技术路径。
Abstract:
To address the issue of insufficient automatic classification accuracy in ethnic costume images due to complex textures and diverse features, posing a significant challenge to their digital preservation and cultural transmission. An improved ResNet-50 model was built by integrating a convolutional attention mechanism into a residual network framework, in order to boost the model’s capability in capturing distinctive features from ethnic costume images and improve its generalization performance in classification tasks. Based on a dataset comprising 15 000 images across five categories of ethnic costumes of Han, Miao, Mongolian, Manchu, and Tibetan, the performance of the proposed model was compared with that of other models including ResNet-34, ResNet-50, VGG16, and AlexNet. The results indicate that the improved model significantly outperforms the others across multiple evaluation metrics, such as accuracy, precision, recall, and F1-score. It effectively enhances the automated classification ability for ethnic costume images and offers an efficient and reliable technical approach for supporting the digital preservation and transmission of ethnic costumes.

参考文献/References:

[1] 中共中央办公厅,国务院办公厅.关于进一步加强非物质文化遗产保护工作的意见[EB/OL].(2021-08-12)[2024-07-07]. https://www.gov.cn/zhengce/2021-08/12/content_5630974.html.
[2] 中共中央办公厅,国务院办公厅.国家“十四五”文化发展规划[EB/OL].(2022-08-16)[2024-07-07]. https://www.gov.cn/zhengce/2022-08/16/content_5705612.html.
[3] 中国纺织工业联合会. “十四五”纺织非物质文化遗产工作行业性指导意见[EB/OL].(2021-12-28)[2024-07-07]. https://www.cntac.org.cn/guihua/202112/t20211228_4229927.html.
[4] SUGANYADEVI S, SEETHALAKSHMI V, BALASAMY K. A review on deep learning in medical image analysis[J]. International Journal of Multimedia Information Retrieval, 2022, 11(1): 19-38.
[5] FRACAROLLI J A, PAVARIN F F A, CASTRO W, et al. Computer vision applied to food and agricultural products[J]. Revista Ciência Agron?mica, 2020, 51(5): e20207749.
[6] ANSARI S, NAGHDY F, DU H P, et al. Driver mental fatigue detection based on head posture using new modified ReLU-BiLSTM deep neural network[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(8): 10957-10969.
[7] BADUGE S K, THILAKARATHNA S, PERERA J S, et al. Artificial intelligence and smart vision for building and construction 4.0: machine and deep learning methods and applications[J]. Automation in Construction, 2022, 141: 104440.
[8] 韩曙光, 姜凯文, 赵丽妍. 基于深度学习的服装三要素识别[J]. 服装学报, 2022, 7(5): 399- 407.
HAN Shuguang, JIANG Kaiwen, ZHAO Liyan. Recognition of clothing "three elements" based on deep learning[J]. Journal of Clothing Research, 2022, 7(5): 399- 407.(in Chinese)
[9] GENG Z M, LIN B J, YUAN Y, et al. Garment fabric pattern classification via ResNet-34[C]//2023 International Conference on Power Communication Computing and Networking Technologies. Wuhan:ACM, 2023: 1- 4.
[10] 杨惠珺, 陈鹤文. 基于GoogLeNet的秦绣纹样的识别与分类研究[J]. 家具与室内装饰, 2023, 30(5): 38- 42.
YANG Huijun, CHEN Hewen. Recognition and classification of Qin embroidery patterns based on GoogLeNet[J]. Furniture and Interior Design, 2023, 30(5): 38- 42.(in Chinese)
[11] 吴圣美, 刘骊, 付晓东, 等. 结合人体检测和多任务学习的少数民族服装识别[J]. 中国图象图形学报, 2019, 24(4): 562-572.
WU Shengmei, LIU Li, FU Xiaodong, et al. Human detection and multi-task learning for minority clothing recognition[J]. Journal of Image and Graphics, 2019, 24(4): 562-572.(in Chinese)
[12] 候红涛, 汪威, 申红婷, 等. 基于自适应图像增强和CNN的少数民族服饰图像分类研究[J]. 现代计算机, 2022(24): 29-35.
HOU Hongtao, WANG Wei, SHEN Hongting, et al. Research on minority clothing classification based on adaptive image enhancement and convolutional neural network[J]. Modern Computer, 2022(24): 29-35.(in Chinese)
[13] 杨冰, 徐丹, 张豪远, 等. 基于改进的DenseNet-BC对少数民族服饰的识别[J]. 浙江大学学报(理学版), 2021, 48(6): 676- 683.
YANG Bing, XU Dan, ZHANG Haoyuan, et al. Minority clothing recognition based on improved DenseNet-BC[J]. Journal of Zhejiang University(Science Edition), 2021, 48(6): 676- 683.(in Chinese)
[14] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 770-778.
[15] MNIH V, HEESS N M O, GRAVES A, et al. Recurrent models of visual attention[C]//2014 International Conference on Neural Information Processing Systems. Montreal: NIPS, 2014: 2204-2212.
[16] HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 7132-7141.
[17] 耿增民, 商书元, 邵新艳, 等. 基于层次语义的Web服装图像智能采集方法[J]. 计算机科学, 2016, 43(增刊2): 252-255.
GENG Zengmin, SHANG Shuyuan, SHAO Xinyan, et al. Intelligent collection method for Web clothing images based on hierar-chical semantics[J]. Computer Science, 2016, 43(Sup 2): 252-255.(in Chinese)
[18] 赵露露, 沈玲, 洪日昌. 图像修复研究进展综述[J]. 计算机科学, 2021, 48(3): 14-26.
ZHAO Lulu, SHEN Ling, HONG Richang. Survey on image inpainting research progress[J]. Computer Science, 2021, 48(3): 14-26.(in Chinese)
[19] 郭玥秀, 杨伟, 刘琦, 等. 残差网络研究综述[J]. 计算机应用研究, 2020, 37(5): 1292-1297.
GUO Yuexiu, YANG Wei, LIU Qi, et al. Survey of residual network[J]. Application Research of Computers, 2020, 37(5): 1292-1297.(in Chinese)
[20] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Computer Vision—ECCV 2018. Cham: Springer, 2018: 3-19.
[21] HETTIHEWA K, KOBCHAISAWAT T, TANPOWPONG N, et al. MANet: a multi-attention network for automatic liver tumor segmentation in computed tomography(CT)imaging[J]. Scientific Reports, 2023, 13: 20098.
(责任编辑:沈天琦)

更新日期/Last Update: 2025-12-30