Abstract
In China, the damage of ancient Yi books are serious. Due to the lack of ancient Yi experts, the repairation of ancient Yi books is progressing very slowly. The artificial intelligence is successful in the field of image and text, so it is feasible for the automatic restoration of ancient books. In this article, a generative adversarial networks with dual discriminator (DDGAN) is designed to restore incomplete characters in the ancient Yi literature. The DDGAN integrates the deep convolution generative adversarial network with an ancient Yi comparison discriminator. Through two training stages, it could iteratively optimizes the ancient Yi character generation networks to obtain the text generator According to the loss of comparison discriminator, DDGAN mode could be optimized. The DDGAN model can generate characters to restore the missing stroke in the ancient Yi. The experiment shows that the proposed method achieves a restoration rate of 77.3% when no more than one third of the characters are missing. This work is effective for the protection of Yi ancient books.
- [1] Fazil Altinel, Mete Ozay, and Takayuki Okatani. 2018. Deep structured energy-based image inpainting. In 2018 24th International Conference on Pattern Recognition (ICPR’18). IEEE, Beijing, China, 423–428. Google Scholar
Cross Ref
- [2] . 2017. SegNet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 39, 12 (2017), 2481–2495. Google Scholar
Cross Ref
- [3] . 2019. Grayification: A meaningful grayscale conversion to improve handwritten historical documents analysis. Pattern Recognition Letters 121 (2019), 46–51.Google Scholar
Digital Library
- [4] . 2018. Hyperspectral image classification with Markov random fields and a convolutional neural network. IEEE Transactions on Image Processing 27, 5 (2018), 2354–2367. Google Scholar
Digital Library
- [5] Bo Chang, Qiong Zhang, Shenyi Pan, and Lili Meng. 2018. Generating handwritten Chinese characters using CycleGAN. In IEEE Winter Conference on Applications of Computer Vision (WACV’18). IEEE, 199–207. Google Scholar
Cross Ref
- [6] Jie Chang, Yujun Gu, Ya Zhang, Yan-Feng Wang, and CM Innovation. 2018. Chinese handwriting imitation with hierarchical generative adversarial network. In BMVC. Springer, Northumbria University, United Kingdom, 290.Google Scholar
- [7] . 2010. Occluded text restoration and recognition. In Proceedings of the 9th IAPR International Workshop on Document Analysis Systems. 151–158.Google Scholar
Digital Library
- [8] . 2019. A recognition method of ancient Yi script based on deep learning. International Journal of Computer and Information Engineering 13, 9 (2019), 508–515.Google Scholar
- [9] Dan C. Cireşan, Ueli Meier, and Jürgen Schmidhuber. 2012. Transfer learning for Latin and Chinese characters with deep neural networks. In International Joint Conference on Neural Networks (IJCNN’12). Brisbane, Australia. Google Scholar
Cross Ref
- [10] . 2004. Region filling and object removal by exemplar-based image inpainting. IEEE Transactions on Image Processing 13, 9 (2004), 1200–1212. Google Scholar
Digital Library
- [11] . 2017. Unsupervised image-to-image translation with generative adversarial networks. arXiv:1701.02676.Google Scholar
- [12] . 2020. Image inpainting: A review. Neural Processing Letters 51, 2 (2020), 2007–2028. Google Scholar
Digital Library
- [13] . 2020. Generative adversarial networks. Communications of the ACM 63, 11 (2020), 139–144. Google Scholar
Digital Library
- [14] Paul Harrison. 2001. A non-hierarchical procedure for re-synthesis of complex textures. In Proceedings of the International Conferences in Central Europe on Computer Graphics, Visualization and Computer Vision. University of West Bohemia, Plzen, Czech Republic, 190–197.Google Scholar
- [15] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’16). IEEE, Nevada Las Vegas, USA, 770–778. Google Scholar
Cross Ref
- [16] . 2019. Pairwise-comparison-based rank learning for benchmarking image restoration algorithms. IEEE Transactions on Multimedia 21, 8 (2019), 2042–2056. Google Scholar
Cross Ref
- [17] . 2020. FormNet: Formatted learning for image restoration. IEEE Transactions on Image Processing 29 (2020), 6302–6314. Google Scholar
Cross Ref
- [18] Shi Jinbo. 2018. A modest study and arrangements on contemporary minority nationalities characters historical documents. Qinghai Nationalities Research 29, 3 (2018), 108–115.Google Scholar
- [19] . 2016. Perceptual losses for real-time style transfer and super-resolution. In European Conference on Computer Vision. Springer, 694–711. Google Scholar
Cross Ref
- [20] . 2017. Improving consistency and correctness of sequence inpainting using semantically guided generative adversarial network. arXiv:1711.06106.Google Scholar
- [21] . 2018. On the convergence of competitive, multi-agent gradient-based learning. arXiv:1804.05464.Google Scholar
- [22] R. Memisevic. 2011. Gradient-based learning of higher-order image features. In the 2011 International Conference on Computer Vision (ICCV). IEEE, Barcelona, Spain, 1591–1598. Google Scholar
Digital Library
- [23] . 2019. From night to day: GANs based low quality image enhancement. Neural Processing Letters 50, 1 (2019), 799–814. Google Scholar
Digital Library
- [24] . 2014. Conditional generative adversarial nets. arXiv:1411.1784.Google Scholar
- [25] Deepak Pathak, Philipp Krahenbuhl, Jeff Donahue, Trevor Darrell, and Alexei A. Efros. 2016. Context encoders: Feature learning by inpainting. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, Nevada Las Vegas, USA, 2536–2544. Google Scholar
Cross Ref
- [26] . 2015. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv:1511.06434.Google Scholar
- [27] . 2003. Best practices for convolutional neural networks applied to visual document analysis. In ICDAR, Vol. 3.Google Scholar
- [28] . 2020. SP-GAN: Self-growing and pruning generative adversarial networks. IEEE Transactions on Neural Networks and Learning Systems 32, 6 (2020), 2458–2469. Google Scholar
Cross Ref
- [29] Aaron Van Oord, Nal Kalchbrenner, and Koray Kavukcuoglu. 2016. Pixel recurrent neural networks. In International Conference on Machine Learning (ICML). PMLR, New York, USA, 1747–1756.Google Scholar
- [30] Yuping L. U. W. U. Xie and Wang M. 2014. Construction on character set of ancient Yi in Guizhou. Journal of Chinese Information Processing. 28, 4 (2014). Google Scholar
Cross Ref
- [31] Wang F. 2003. Tectonic-lithofacies palaeogeography of the Silurian in Sichuan-Yunnan-Guizhou-Guangxi region. Journal of Palaeogeography 5 (2003), 180–186. Google Scholar
Cross Ref
- [32] Wang Z. 2022. A brief account of the Yis’ character inscribed on metals and rocks. Guizhou Etmnic Studies 2 (2002), 156–163.Google Scholar
- [33] . 1996. Linking broken character borders with variable sized masks to improve recognition. Pattern Recognition 29, 8 (1996), 1429–1435. Google Scholar
Cross Ref
- [34] . 2016. Semantic image inpainting with perceptual and contextual losses. arXiv:1607.07539 2, 3.Google Scholar
- [35] Raymond A. Yeh, Chen Chen, Teck Yian Lim, Alexander G. Schwing, Mark Hasegawa-Johnson, and Minh N. Do. 2017. Semantic image inpainting with deep generative models. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, Nevada Las Vegas, USA, 5485–5493. Google Scholar
Cross Ref
- [36] Donggeun Yoo, Namil Kim, Sunggyun Park, Anthony S. Paek, and In So Kweon. 2016. Pixel-level domain transfer. In European Conference on Computer Vision. (ECCV). Springer, Drente Amsterdam, Netherlands, 517–532. Google Scholar
Cross Ref
- [37] . 2001. Reconstruction of broken handwritten digits based on structural morphological features. Pattern Recognition 34, 2 (2001), 235–254. Google Scholar
Cross Ref
- [38] Richard Zhang, Phillip Isola, and Alexei A. Efros. 2016. Colorful image colorization. In European Conference on Computer Vision (ECCV’16). Springer, Drente Amsterdam, Netherlands, 649–666. Google Scholar
Cross Ref
- [39] . 2019. Generative adversarial network-based intra prediction for video coding. IEEE Transactions on Multimedia 22, 1 (2019), 45–58. Google Scholar
Digital Library
Index Terms
Dual Discriminator GAN: Restoring Ancient Yi Characters
Recommendations
DeepNetDevanagari: a deep learning model for Devanagari ancient character recognition
AbstractDevanagari script is the most widely used script in India and other Asian countries. There is a rich collection of ancient Devanagari manuscripts, which is a wealth of knowledge. To make these manuscripts available to people, efforts are being ...
HI-GAN: A hierarchical generative adversarial network for blind denoising of real photographs
AbstractAlthough deep convolutional neural networks (DCNNs) and generative adversarial networks (GANs) have achieved remarkable success in image denoising, they have been facing a severe problem of the trade-off between removing noise and ...
Identification of Ancient Tamil Letters and Its Characters: Automatic Date Fixation Based on Contour-let Technique
ICGSP '17: Proceedings of the 1st International Conference on Graphics and Signal ProcessingAncient Tamil letters and its hand written characters appears in the temple inscriptions are difficult. Besides the ancient Tamil letters and its characters are different from the modern Tamil letters by its character. This paper focuses on the ...






Comments