人工智能目标检测技术在书画文物病害调查中的应用

邓旭帅; 李子璇; 张云春; 沐蕊

doi:10.16152/j.cnki.xdxbzr.2025-01-008

您当前的位置：

首页 >

文章列表页 >

人工智能目标检测技术在书画文物病害调查中的应用

文化遗产数字化保护 | 更新时间：2025-02-08

- 人工智能目标检测技术在书画文物病害调查中的应用
- Application of artificial intelligence object detection technology in disease identification of calligraphy and painting cultural relics
- 西北大学学报（自然科学版） 2025年55卷第1期页码：98-105
- 作者机构：
  
  1.云南大学　历史与档案学院，云南　昆明　 650091
  2.云南大学　软件学院，云南　昆明　 650504
  3.云南省博物馆，云南　昆明　 650214
- 作者简介：
  
  邓旭帅，男，从事文物保护研究，2829860757@qq.com。
  张云春，男，博士，副教授，从事人工智能应用研究，yczhang@ynu.edu.cn。
- 基金信息：
  
  国家档案局科技项目计划(2021-B-03;2022-B-002)
- DOI：10.16152/j.cnki.xdxbzr.2025-01-008
  中图分类号： K854.3
- 纸质出版日期：2025-02-25，
  
  收稿日期：2024-04-20，
- 稿件说明：
移动端阅览
邓旭帅, 李子璇, 张云春, 等. 人工智能目标检测技术在书画文物病害调查中的应用[J]. 西北大学学报（自然科学版）, 2025,55(1):98-105.

DENG XUSHUAI, LI ZIXUAN, ZHANG YUNCHUN, et al. Application of artificial intelligence object detection technology in disease identification of calligraphy and painting cultural relics. [J]. Journal of northwest university (natural science edition), 2025, 55(1): 98-105.
邓旭帅, 李子璇, 张云春, 等. 人工智能目标检测技术在书画文物病害调查中的应用[J]. 西北大学学报（自然科学版）, 2025,55(1):98-105. DOI： 10.16152/j.cnki.xdxbzr.2025-01-008.

DENG XUSHUAI, LI ZIXUAN, ZHANG YUNCHUN, et al. Application of artificial intelligence object detection technology in disease identification of calligraphy and painting cultural relics. [J]. Journal of northwest university (natural science edition), 2025, 55(1): 98-105. DOI： 10.16152/j.cnki.xdxbzr.2025-01-008.

摘要

针对书画文物保护工作中人工病害调查和病害图绘制效率低的问题，探索了基于深度神经网络的目标检测技术识别书画病害的可行性。选择YOLOv5系列模型并根据本研究任务特点对其结构做了优化，包括FGSM算法、CmBN策略、Dropblock正则化和CIOU_Loss损失函数。利用博物馆馆藏书画文物素材，融合Mosaic数据增强方法进行书画文物图片的增强，设计了滑动窗口检测技术、图像逐层分析和定位裁剪技术，初步训练出了2个具备病害识别功能的模型，根据模型性能检验指标最终选择了YOLOv5x6作为本研究任务的模型。测试结果表明，该模型以较高的准确率和查全率识别出了待检测病害，用时仅为人工的千分之一。该技术的引入可极大提高文物病害识别效率，并且在病害识别过程中保持客观、稳定的标准。

Abstract

Targeting the low-efficiency problem of manual disease identification and disease mapping in the protection of calligraphy and painting cultural relics

this paper explores the feasibility of deep neural network-based object detection technology to identify calligraphy and painting diseases. We design a series of YOLOv5 models with some architecture optimizations based on the special requirements of disease identification. The optimizations include FGSM algorithm

CmBN strategy

Dropblock normalization

and CIOU_Loss loss function. Using the materials of calligraphy and painting of cultural relics in the museum as inputs

we enhance the images by combining Mosaic data enhancement method. Two disease identification deep learning models are trained based on some improvements

including sliding-window detection

image clipping based on layer-by-layer image analysis and positioning

etc. By evaluating the models with bench-mark performance metrics

this paper chooses YOLOv5x6 for our task. The experimental results show that YOLOv5x6 outperforms the other models with the best precision and recall. This model takes one-thousandth time compared with manual work. The introduction of deep learning techniques in disease identification not only helps to improve the efficiency of disease identification of cultural relics

but also provides objective and stable standards in the processes of disease identification.

关键词

病害识别目标检测图像处理YOLOv5深度神经网络

Keywords

diseases identificationobject detectionimage processingYOLOv5deep neural networks

references

南京博物院. 中国书画文物修复导则[M]. 南京: 译林出版社, 2017: 12.

田朋, 尚小临, 王若苏. 古书画保护修复中的二次脱色现象及预防[J]. 文物保护与考古科学, 2022, 34(1): 35-41.

TIAN P, SHANG X L, WANG R S. Research on the secondary shedding of pigment in the restoration of ancient paintings and calligraphy and its prevention[J]. Sciences of Conservation and Archaeology, 2022, 34(1): 35-41.

王婕, 巨建伟, 王天凤, 等. 唐卡病害图绘制的规范化研究：以《威罗瓦金刚画像轴》病害图的绘制为例[J]. 中国文物科学研究, 2020(4): 81-88.

WANG J, JU J W, WANG T F, et al. A study on the drawing Tangka disease maps: Taking the drawing of disease maps in the "Verois Vajra Thangka " as an example[J]. China Cultural Heritage Scientific Research, 2020(4): 81-88.

国家文物局. 馆藏纸质文物病害分类与图示：WWT 0026-2010[S]. 北京: 中国标准出版社, 2010.

谷明岩. 基于改进U-Net的壁画颜料层脱落病害提取研究[D]. 北京: 北京建筑大学, 2020.

刘晓琴, 侯妙乐, 董友强, 等. 基于高光谱影像的瞿昙寺壁画颜料层脱落病害评估[J]. 地理信息世界, 2019, 26(5): 22-28.

LIU X Q, HOU M L, DONG Y Q, et al. Extraction and evaluation of the disease of the mural paint Loss[J]. Geomatics World, 2019, 26(5): 22-28.

胡春梅, 王瑜, 黄浩雯, 等. 基于纹理的文物病害高精度自动提取方法: CN110796181B[P]. 2022-05-03.

曹建芳, 李艳飞, 崔红艳, 等. 改进的区域生长算法在寺观壁画脱落病害标定中的应用[J]. 新疆大学学报(自然科学版), 2018, 35(4): 429-436.

CAO J F, LI Y F, CUI H Y, et al. The application of improved region growing algorithm for the automatic calibration of shedding disease on temple murals[J]. Journal of Xinjiang University (Natural Science Edition), 2018, 35(4): 429-436.

张楠, 张乾, 冯伟, 等. 古代壁画病害标识系统及其在敦煌莫高窟的应用[J]. 敦煌研究, 2017(2): 135-140.

ZHANG N, ZHANG Q, FENG W, et al. The deterioration identification system of ancient murals and its application in the Dunhuang Mogao Grottoes[J]. Dunhuang Research, 2017(2): 135-140.

王珺, 孙进越, 俞凯, 等. 基于分组LSTM与CNN的青铜器锈蚀类别智能标识方法[J]. 西北大学学报(自然科学版), 2021, 51(5): 778-786.

WANG J, SUN J Y, YU K, et al. Intelligent identification method of bronze rust category based on grouping LSTM and CNN[J]. Journal of Northwest University (Natural Science Edition), 2021, 51(5): 778-786.

耿国华, 冯龙, 李康, 等. 秦陵文物数字化及虚拟复原研究综述[J]. 西北大学学报(自然科学版), 2021, 51(5): 710-721.

GENG G H, FENG L, LI K, et al. A literature review on the digitization and virtual restoration of cultural relics in the Emperor Qinshihuang's Mausoleum[J]. Journal of Northwest University (Natural Science Edition), 2021, 51(5): 710-721.

李彩艳, 王慧琴, 吴萌, 等. 唐墓室壁画泥斑病害自动标定及虚拟修复[J]. 计算机工程与应用, 2016, 52(15): 233-236.

LI C Y, WANG H Q, WU M, et al. Automatic recognition and virtual restoration of mud spot disease of Tang dynasty tomb murals image[J]. Computer Engineering and Applications, 2016, 52(15): 233-236.

姜军, 卓嘎, 王朝霞. 西藏数字壁画泥斑病害自动标定修复方法仿真[J]. 计算机仿真, 2018, 35(11): 215-219.

JIANG J, ZHUO G, WANG Z X. Digital curtain diameter mould disease auto calibration and restoration method simulation[J]. Computer Simulation, 2018, 35(11): 215-219.

谢富, 朱定局. 深度学习目标检测方法综述[J]. 计算机系统应用, 2022, 31(2): 1-12.

XIE F, ZHU D J. Survey on deep learning object detection[J]. Computer Systems & Applications, 2022, 31(2): 1-12.

谭营. 人工智能知识讲座[M]. 北京: 人民出版社, 2018.

NARESH K, SATISH G N, RAM K B, et al. Machine learning based recognition of crops diseases by CNN[J]. International Journal of Innovative Technology and Exploring Engineering, 2019, 8(9): 1264-1268.

刘学增, 桑运龙, 苏云帆. 基于数字图像处理的隧道渗漏水病害检测技术[J]. 岩石力学与工程学报, 2012, 31(S2): 3779-3786.

LIU X Z, SANG Y L, SU Y F. Detection technology of tunnel leakage disaster based on digital image processing[J]. Chinese Journal of Rock Mechanics and Engineering, 2012, 31(S2): 3779-3786.

XIANG K, PENG L L, YANG H Q, et al. A novel weight pruning strategy for light weight neural networks with application to the diagnosis of skin disease[J]. Applied Soft Computing, 2021, 111: 107707.

ZHENG Z H, WANG P, LIU J W, et al. Distance-IoU loss: Faster and better learning for bounding box regression[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(7): 12993-13000.

冯增木. 中国书画装裱[M]. 3版. 济南: 山东科学技术出版社, 2002: 207.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

面向甲骨文目标检测的大规模数据集生成技术

基于可变形卷积的小肠间质瘤多模型联合检测

计算机视觉辅助训练系统研究——以射击类项目为例

一种基于计算机视觉的单目测量头盔转动角度的方法

基于映射的图像多角度复小波变换