[1]谢标峰,陈首虹,黄吉祥,等.基于卷积神经网络的石材镶嵌工艺颜色匹配方法[J].华侨大学学报(自然科学版),2025,46(4):393-399.[doi:10.11830/ISSN.1000-5013.202503023]
　XIE Biaofeng,CHEN Shouhong,HUANG Jixiang,et al.Color Matching Method for Stone Tessellation Process Based on Convolutional Neural Network[J].Journal of Huaqiao University(Natural Science),2025,46(4):393-399.[doi:10.11830/ISSN.1000-5013.202503023]
点击复制

基于卷积神经网络的石材镶嵌工艺颜色匹配方法()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第46卷
期数:
2025年第4期
页码:
393-399
栏目:
出版日期:
2025-07-16

文章信息/Info

Title:
Color Matching Method for Stone Tessellation Process Based on Convolutional Neural Network
文章编号:
1000-5013(2025)04-0393-07
作者:
谢标峰1; 2;  陈首虹3;  黄吉祥1; 2;  李建新1; 2;  黄身桂1; 2
1. 华侨大学 制造工程研究院, 福建 厦门 361021; 2. 南安华大石材产业技术研究院, 福建 南安 362342;3. 华侨大学 机电及自动化学院, 福建 厦门 361021
Author(s):
XIE Biaofeng1; 2;  CHEN Shouhong3;  HUANG Jixiang1; 2;  LI Jianxin1; 2;  HUANG Shengui1; 2
1. Institute of Manufacturing Engineering, Huaqiao University, Xiamen 361021, China; 2. Nan’an-HQU Institute of Stone Industry Innovations Technology, Quanzhou 362342, China; 3. College of Mechanical Engineering and Automation, Huaqiao University, Xiamen 361021, China
关键词:
镶嵌工艺;  石材分类;  卷积神经网络;  工艺优化
Keywords:
tessellation process;  stone classification;  convolutional neural network;  process optimization
分类号:
TP399;TU564.2
DOI:
10.11830/ISSN.1000-5013.202503023
文献标志码:
A
摘要:
针对石材镶嵌工艺中人工选料耗时长、得到产品质量不稳定的问题,提出一种基于卷积神经网络模型的石材镶嵌工艺颜色匹配方法。对生产车间的石材扫描样本预处理构建石材图像数据集;训练不同卷积神经网络,筛选出在数据集上分类效果最好的石材分类模型;为验证其分类效果,提取目标图像的颜色区域作为输入进行实际生产。结果表明:采用所提方法匹配的石材生产出的镶嵌产品与目标图像在颜色一致性方面表现优异,视觉感知效果高度接近;所提方法既能提高石材的挑选效率,又保障了产品质量的稳定性。
Abstract:
To address the problem of time-consuming manual material selection and unstable product quality in the stone tessellation process, a color matching method based on convolutional neural network model is proposed. Stone samples scanned from the production workshop were preprocessed to construct the stone image dataset. Different convolutional neural networks were trained to screen out the best stone classification model. In order to validate the model’s effectiveness, the color region of the target image was extracted and used as the input for actual production. The results demonstrated that the products produced using the proposed method were excellent in terms of color consistency, and the visual perception effect was highly similar. The proposed method significantly improve the efficiency of stone selection and ensure the stability of product quality.

参考文献/References:

[1] 熊瑛子.石材镶嵌艺术在中国当代壁画创作中的运用[J].天工,2022,36(1):90-92.DOI:10.3969/j.issn.2095-7556.2022.01.024
[2] 熊瑛子.石材镶嵌艺术风格研究[J].美术文献,2019(5):9-11.DOI:10.16585/j.cnki.mswx.2019.05.005.
[3] 熊瑛子.石材镶嵌艺术材料与创作手法分析[J].艺术品鉴,2019,15(12):294-295.
[4] 陈瑶.中国传统家具镶嵌艺术及现代化技术的研究[D].长沙:中南林业科技大学,2006.
[5] 陈韫如.清代宫廷家具中的镶嵌工艺[J].文物鉴定与鉴赏,2022(8):116-119.DOI:10.20005/j.cnki.issn.1674-8697.2022.08.
[6] 李佳明.传统镶嵌工艺在当代漆艺中的运用[D].沈阳:鲁迅美术学院,2019.
[7] 杜金华.基于颜色特征和逻辑回归的饰面花岗石图像识别技术研究[D].厦门:华侨大学,2018.
[8] 康利娟.基于图像颜色的石材分类算法及测试平台研究与实现[D].武汉:武汉理工大学,2009.
[9] 杨杰,杨静宜.基于颜色直方图的石材图像检索[J].武汉理工大学学报(信息与管理工程版),2009,31(2):173-176.DOI:10.3963/j.issn.1007-144X.2009.02.001.
[10] 黄士真,耿栋.基于神经网络的建筑装饰石材智能分类研究: 以天然大理石分类为例[J].广西城镇建设,2023(2):90-97.DOI:10.3969/j.issn.1672-7045.2023.02.011.
[11] AGRAWAL N,GOVIL H.A deep residual convolutional neural network for mineral classification[J].Advances in Space Research,2023,71(8):3186-3202.DOI:10.1016/j.asr.2022.12.028.
[12] TROPEA M,FEDELE G,DE LUCA R,et al.Automatic stones classification through a CNN-based approach[J].Sensors,2022,22(16):6292.DOI:10.3390/s22166292.
[13] 刘晨,赵晓晖,梁乃川,等.基于ResNet50和迁移学习的岩性识别与分类研究[J].计算机与数字工程,2021,49(12):2526-2530.DOI:10.3969/j.issn.1672-9722.2021.12.020.
[14] 程志清.基于深度学习的工业机器人视觉引导方法研究[J].信息记录材料,2024,25(12):110-112.DOI:10.16009/j.cnki.cn13-1295/tq.2024.12.048.
[15] 檀为龙.深度强化学习在智能制造中的创新应用与趋势展望[J].中国战略新兴产业,2024(33):161-163.DOI:10.3778/j.issn.1002-8331.2008-0431.
[16] XIE Saining,GIRSHICK R,DOLLáR P,et al.Aggregated residual transformations for deep neural networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu:IEEE Press,2017:5987-5995.DOI:10.1109/CVPR.2017.634.
[17] BOUDIAF A,BENLAHMIDI S,DAHANE A,et al.Development of hybrid models based on alexnet and machine learning approaches for strip steel surface defect classification[J].Journal of Failure Analysis and Prevention,2024,24(3):1376-1394.DOI:10.1007/s11668-024-01927-5.
[18] WANG Wei,ZHANG Chengwen,TIAN Jinge,et al.High-resolution radar target recognition via inception-based VGG(IVGG)networks[J].Computational Intelligence and Neuroscience,2020,2020(1):8893419.DOI:10.1155/2020/8893419.
[19] MAHAUR B,MISHRA K K,SINGH N.Improved residual network based on norm-preservation for visual recognition[J].Neural Networks,2023,157:305-322.DOI:10.1016/j.neunet.2022.10.023.

备注/Memo

备注/Memo:
收稿日期: 2025-03-12
通信作者: 黄身桂(1981-),男,副教授,博士,主要从事智能制造与高效精密加工、机器人磨抛加工、智能制造装备设计与开发的研究。E-mail:shghuang@hqu.edu.cn。
基金项目: 福建省科技计划资助项目(2022H0018)
更新日期/Last Update: 2025-07-20