A Gaussian Mixture Model Approach for Robust Watermark Text Detection in Documents

Authors

  • Maheshwari S Hiremath Department of Studies in Computer Science, Davanagere University, Davangere, India
  • Basavanna Mahadevappa Department of Studies in Computer Science, Davanagere University, Davangere, India
  • Shivanand Sharanappa Gornale School of Mathematics and Computing Sciences. Rani Channamma University, Belagavi, India

DOI:

https://doi.org/10.14738/tmlai.1305.19505

Keywords:

Watermark text, Gaussian Mixture Model, Canny edge, Morphological operation, Otsu Thresholding

Abstract

Watermark text detection is a crucial process in image processing and digital forensics, particularly for ensuring content authenticity and preventing unauthorized use of digital media. Watermarks, often embedded in images, serve as a visible or invisible form of protection, and their detection is essential for verifying the integrity and ownership of digital assets. Detecting watermark text which is often subtle, semi-transparent, or integrated into the background presents significant challenges. In this paper, we propose an effective approach for watermark text detection using a Gaussian Mixture Model (GMM), an edge detection technique, and morphological functions. A Gaussian Mixture Model (GMM) with four cluster components is applied to model the distribution of pixel intensities. The GMM represents the intensity histogram as a mixture of Gaussian distributions, each parameterized by its mean, variance, and weight. Canny edge detection is applied to the grayscale image to retain the actual edge information. The resulting image is then refined using morphological closing a process that involves dilation followed by erosion to fill small gaps and smooth edge contours. To further isolate meaningful edges, Otsu’s thresholding method is applied, selecting an optimal threshold that minimizes intra-class variance between the foreground and background. Finally, the watermark text is embedded into the image by modifying the pixel values in a predefined region, resulting in a watermarked output image.

References

Gong, Z., Qin, N. and Zhang, G., 2021. Visible watermarking in document images using two-stage fuzzy inference system. The Visual Computer, pp.1-12.

Singh, B. and Sharma, M.K., 2021. Tamper detection technique for document images using zero watermarking in wavelet domain. Computers & Electrical Engineering, 89, p.106925.

Laouamer, L. and Tayan, O., 2018. Performance evaluation of a document image watermarking approach with enhanced tamper localization and recovery. IEEE Access, 6, pp.26144-26166.

Tan, L., Hu, K., Zhou, X., Chen, R. and Jiang, W., 2019. Print-scan invariant text image watermarking for hardcopy document authentication. Multimedia Tools and Applications, 78(10), pp.13189-13211.

Zenati, A., Ouarda, W. and Alimi, A.M., 2021. A new digital steganography system based on hiding online signature within document image data in YUV color space. Multimedia Tools and Applications, 80(12), pp.18653-18676.

J. T. Brassil, S. Low, N. F. Maxemchuk, and L. O’Gorman, “Electronic Marking and Identification Techniques to Discourage Document Copying,” IEEE Journal on Selected Areas in Communications, vol. 13, no. 8, October 1995, pp. 1495-1504.

J. T. Brassil, S. Low, and N. F. Maxemchuk, "Copyright Protection for the Electronic Distribution of Text Documents," Proceedings of the IEEE, vol. 87, no. 7, July 1999, pp.1181-1196.

N. F. Maxemchuk, S. H. Low, “Performance Comparison of Two Text Marking Methods,” IEEE Journal of Selected Areas in Communications (JSAC), May 1998. vol. 16 no. 4 1998. pp. 561-572.

N. F. Maxemchuk, “Electronic Document Distribution,” AT&T Technical Journal, September 1994, pp. 73-80. 6.

N. F. Maxemchuk and S. Low, “Marking Text Documents,” Proceedings of the IEEE International Conference on Image Processing, Washington, DC, Oct. 26-29, 1997, pp. 13-16.

S. H. Low, N. F. Maxemchuk, and A. M. Lapone, "Document Identification for Copyright Protection Using Centroid Detection," IEEE Transactions on Communications, Mar. 1998, vol. 46, no.3, pp 372-381.

D. Huang and H. Yan, "Interword distance changes represented by sine waves for watermarking text images," IEEE Trans. Circuits and Systems for Video Technology, Vol.11, No.12, pp.1237-1245, Dec 2001.

T. Amano and D. Misaki, "A feature calibration method for watermarking of document images," Proceedings of ICDAR, pp.91-94, 1999.

Fahmy, G. (2009, December). A quasi-blind watermark extraction of watermarked natural preserve transform images on a DSP board. In 2009 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT) (pp. 259-264). IEEE.

Palit, S., and Garain, U. (2006, April). A novel technique for the watermarking of symbolically compressed documents. In Second International Conference on Document Image Analysis for Libraries (DIAL'06) (pp. 6-pp). IEEE.

TamilSelvan, R., Prathap, I., Ramalingam, A., and Raghavan, S. (2009, June). A novel approach to watermark text documents based on Eigen values. In 2009 International Conference on Network and Service Security (pp. 1-5). IEEE.

Tang, X. H., and Wang, L. N. Text watermarking algorithm based on the stroke of Chinese characters. In 2011 International Conference on Multimedia Technology, (2011, July), (pp. 794-796). IEEE.

Barlaskar, S. A., Singh, S. V., and Laskar, R. H. (2022). Genetic algorithm based optimized watermarking technique using hybrid DCNN-SVR and statistical approach for watermark extraction. Multimedia Tools and Applications, 81(5), 7461-7500.

W.H. Kim, J.U. Hou, S.M. Mun, and H.K. Lee, Convolutional neural network architecture for recovering watermark synchronization. arXiv preprint arXiv:1805.06199, 2018.

H. Fang, D. Chen, Q. Huang, J. Zhang, Z. Ma, W. Zhang, and N. Yu, Deep template based watermarking, IEEE Trans. Circuits Syst. Video Technol. (2020).

Karki, B., Tsai, C. H., Huang, P. C., & Zhong, X. (2024, August). Deep Learning-based Text-in-Image Watermarking. In 2024 IEEE 7th International Conference on Multimedia Information Processing and Retrieval (MIPR) (pp. 376-382). IEEE.

Krubiński, M., Matcovici, S., Grigore, D., Voinea, D., & Popa, A. I. (2024). Watermark text pattern spotting in document images. arXiv preprint arXiv:2401.05167.

Lu, Y., Liu, A., Yu, D., Li, J., & King, I. (2024). An entropy-based text watermarking detection method. arXiv preprint arXiv:2403.13485.

Munyer, T., Tanvir, A. A., Das, A., & Zhong, X. (2024). Deeptextmark: A deep learning-driven text watermarking approach for identifying large language model generated text. IEEE Access, 12, 40508-40520.

Liu, R., Li, S., Li, Z., Sun, Y., Cai, Z., & Dai, D. (2023, October). Research on a watermarking recognition method for PDF documents based on natural language processing. In Fourth International Conference on Computer Science and Communication Technology (ICCSCT 2023) (Vol. 12918, pp. 150-155). SPIE.

Li, L., Zhang, H. J., Meng, J. L., & Lu, Z. M. (2023). Robust pdf watermarking against print–scan attack. Sensors, 23(17), 7365.

Arabi, K., Witter, R. T., Hegde, C., & Cohen, N. (2025). SEAL: Semantic Aware Image Watermarking. arXiv preprint arXiv:2503.12172.

Zhuang Li (2025). BiMarker: Enhancing Text Watermark Detection for Large Language Models with Bipolar Watermarks. arXiv preprint arXiv:2501.12174.

Singh, A., Kumar, R., & Mehta, S. (2018). Rule-based PDF parsing for watermark text detection using the iText library. Journal of Digital Document Processing, 12(3), 145–152.

Chen, Y., Zhang, W., & Liu, J. (2020). CNN-based watermark detection from image-rendered PDF pages. Proceedings of the International Conference on Document Analysis and Recognition (ICDAR). https://doi.org/10.1109/ICDAR.2020.00123.

Liu, Z., Wang, Y., & Zhao, L. (2023). Real-time watermark text detection in PDFs using YOLOv5. Pattern Recognition Letters, 168, 105–113. https://doi.org/10.1016/j.patrec.2023.04.015.

Munyer, T., Tanvir, A., Das, A., & Zhong, X. (2023). DeepTextMark: A deep learning-driven text watermarking approach for identifying large language model generated text. arXiv preprint, arXiv:2305.05773. https://arxiv.org/abs/2305.05773.

Lu, Y., Liu, A., Yu, D., Li, J., & King, I. (2024). An entropy-based text watermarking detection method. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024).

Gu, T., Wang, Z., Huang, K., Yao, Y., Zhang, X., Yang, Y., & Chen, X. (2025). Invisible Entropy: Towards safe and efficient low-entropy LLM watermarking. arXiv preprint, arXiv:2505.14112. https://arxiv.org/abs/2505.14112.

He, C., Wu, D., Zhang, X., & Wu, H. (2024, June). Watermarking text documents with watermarked fonts. In Proceedings of the 2024 ACM workshop on information hiding and multimedia security (pp. 187-197).

Zhou, Y., Zhu, Y., Ye, Q., Qiu, Q., & Jiao, J. (2018). Weakly supervised instance segmentation using class peak response. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 3791-3800).

Bappy, J. H., Roy-Chowdhury, A. K., Bunk, J., Nataraj, L., & Manjunath, B. S. (2017). Exploiting spatial structure for localizing manipulated image regions. In Proceedings of the IEEE international conference on computer vision (pp. 4970-4979).

Amirmazlaghani, M., Rezghi, M., & Amindavar, H. (2015). A novel robust scaling image watermarking scheme based on Gaussian Mixture Model. Expert Systems with Applications, 42(4), 1960-1971.

Downloads

Published

2025-10-28

How to Cite

Hiremath, M. S., Mahadevappa, B., & Gornale, S. S. (2025). A Gaussian Mixture Model Approach for Robust Watermark Text Detection in Documents. Transactions on Engineering and Computing Sciences, 13(05), 87–102. https://doi.org/10.14738/tmlai.1305.19505