基于GAN的印刷微点图像信息恢复算法研究

Bo Yuan, Peng Cao
{"title":"基于GAN的印刷微点图像信息恢复算法研究","authors":"Bo Yuan, Peng Cao","doi":"10.1145/3569966.3571169","DOIUrl":null,"url":null,"abstract":"During printing and shooting, the degradation of printing micro dots significantly affects the decoding and reading of hidden anti-counterfeiting information. However, existing image restoration methods cannot effectively restore image information. Moreover, there are relatively few datasets related to halftone dot images, and most datasets differ from the real data. Therefore, we propose an end-to-end restoration model based on the single-image super-resolution information. Specifically, we constructed a PMD dataset for real printing of anti-counterfeiting scenes. Based on this dataset, we used the high-resolution image information as the target. The positional inclination of the degraded images is corrected using the blank and interline characteristics of the printing micro dots images. The restoration is completed with the help of feature extraction and upsample of ESRGAN. In addition, we propose evaluation measures suitable for error detection, correction, and decoding requirements for microscopic image information. The experimental results show that, within the noise tolerance range, the image information restored by our method has a maximum average bit error rate is 0.97% and a Euclidean distance is 0.00804 pixels, whereas traditional filtering measures cannot effectively restore image information. The experimental results verified the effectiveness and robustness of the proposed method.","PeriodicalId":145580,"journal":{"name":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","volume":"110 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Research on Image Information Restoration Algorithm of Printing Micro Dots Based on GAN\",\"authors\":\"Bo Yuan, Peng Cao\",\"doi\":\"10.1145/3569966.3571169\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"During printing and shooting, the degradation of printing micro dots significantly affects the decoding and reading of hidden anti-counterfeiting information. However, existing image restoration methods cannot effectively restore image information. Moreover, there are relatively few datasets related to halftone dot images, and most datasets differ from the real data. Therefore, we propose an end-to-end restoration model based on the single-image super-resolution information. Specifically, we constructed a PMD dataset for real printing of anti-counterfeiting scenes. Based on this dataset, we used the high-resolution image information as the target. The positional inclination of the degraded images is corrected using the blank and interline characteristics of the printing micro dots images. The restoration is completed with the help of feature extraction and upsample of ESRGAN. In addition, we propose evaluation measures suitable for error detection, correction, and decoding requirements for microscopic image information. The experimental results show that, within the noise tolerance range, the image information restored by our method has a maximum average bit error rate is 0.97% and a Euclidean distance is 0.00804 pixels, whereas traditional filtering measures cannot effectively restore image information. The experimental results verified the effectiveness and robustness of the proposed method.\",\"PeriodicalId\":145580,\"journal\":{\"name\":\"Proceedings of the 5th International Conference on Computer Science and Software Engineering\",\"volume\":\"110 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-10-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 5th International Conference on Computer Science and Software Engineering\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3569966.3571169\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 5th International Conference on Computer Science and Software Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3569966.3571169","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

在印刷和拍摄过程中,印刷微点的退化严重影响隐藏防伪信息的解码和读取。然而,现有的图像恢复方法不能有效地恢复图像信息。此外,与半色调网点图像相关的数据集相对较少,而且大多数数据集与真实数据存在差异。因此,我们提出了一种基于单幅图像超分辨率信息的端到端恢复模型。具体来说,我们构建了一个真实打印防伪场景的PMD数据集。在此数据集的基础上,我们以高分辨率图像信息为目标。利用打印微点图像的空白和行间特性对退化图像的位置倾斜进行校正。利用ESRGAN的特征提取和上采样完成恢复。此外,我们提出了适合于显微图像信息的错误检测、纠错和解码要求的评价措施。实验结果表明,在噪声容限范围内,该方法恢复的图像信息最大平均误码率为0.97%,欧氏距离为0.00804像素,而传统滤波措施无法有效恢复图像信息。实验结果验证了该方法的有效性和鲁棒性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Research on Image Information Restoration Algorithm of Printing Micro Dots Based on GAN
During printing and shooting, the degradation of printing micro dots significantly affects the decoding and reading of hidden anti-counterfeiting information. However, existing image restoration methods cannot effectively restore image information. Moreover, there are relatively few datasets related to halftone dot images, and most datasets differ from the real data. Therefore, we propose an end-to-end restoration model based on the single-image super-resolution information. Specifically, we constructed a PMD dataset for real printing of anti-counterfeiting scenes. Based on this dataset, we used the high-resolution image information as the target. The positional inclination of the degraded images is corrected using the blank and interline characteristics of the printing micro dots images. The restoration is completed with the help of feature extraction and upsample of ESRGAN. In addition, we propose evaluation measures suitable for error detection, correction, and decoding requirements for microscopic image information. The experimental results show that, within the noise tolerance range, the image information restored by our method has a maximum average bit error rate is 0.97% and a Euclidean distance is 0.00804 pixels, whereas traditional filtering measures cannot effectively restore image information. The experimental results verified the effectiveness and robustness of the proposed method.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信