Layout Detection of Punjabi Newspapers using the YOLOv8 Model

doi:10.23940/ijpe.24.03.p7.186193

Abstract

Abstract: Layout analysis is an essential task in the Newspaper recognition system. Conventional techniques for layout analysis, such as top-down and bottom-up methodologies, are environment-specific and cannot achieve accurate results. A novel and potentially effective technique is deep learning-based detection, like the YOLO algorithm. This paper presents a layout analysis of a newspaper using the deep learning model YOLOv8. Newspaper images from different sources are collected and annotated to create the dataset. Around 600 images were collected and annotated. Then, we trained YOLOv8 based on a custom dataset of Punjabi newspapers. We have tested the performance of the trained model over various newspaper images giving a very good accuracy. We have also compared the trained model with a Faster RCNN model with a different backbone. We have even tested the model on other newspapers in other languages.

Key words: YOLOv8, deep learning, layout, newspaper, segmentation, machine vision

Atul Kumar and Gurpreet Singh Lehal. Layout Detection of Punjabi Newspapers using the YOLOv8 Model [J]. Int J Performability Eng, 2024, 20(3): 186-193.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

References

[1] Wong K.Y., Casey R.G., andWahl F.M.Document Analysis System. IBM journal of research and development, vol. 26, no. 6, pp. 647-656, 1982.
[2] Krishnamoorthy M., Nagy G., Seth S., andViswanathan M.Syntactic Segmentation and Labeling of Digitized Pages from Technical Journals. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, no. 7, pp. 737-747, 1993.
[3] Gatos B., Mantzaris S.L., Chandrinos K.V., Tsigris A., andPerantonis S.J.Integrated Algorithms for Newspaper Page Decomposition and Article Tracking. InProceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR'99 (Cat. No. PR00318), IEEE, pp. 559-562, 1999.
[4] Shihab M.I.H., Hasan, M.R., Emon, M.R., Hossen, S.M., Ansary, M.N., Ahmed, I., Rakib, F.R., Dhruvo, S.E., Dip, S.S., Pavel, A.H., and Meghla, M.H. Badlad: A Large Multi-Domain Bengali Document Layout Analysis Dataset. In International Conference on Document Analysis and Recognition, Cham: Springer Nature Switzerland, pp. 326-341, 2023.
[5] Fu P., Zhang X., andYang H.Answer Sheet Layout Analysis Based on YOLOv5s-DC and MSER.The Visual Computer, pp. 1-12, 2023.
[6] Iskandar, N.A.A. Manga Layout Analysis via Deep Learning. In IRC-SET 2022: Proceedings of the 8th IRC Conference on Science, Engineering and Technology, August 2022, Singapore, Singapore: Springer Nature Singapore, pp. 63-73, 2023.
[7] Oliaee, A.H. and Tripp, A.R.Layout Analysis of Historic Architectural Program Documents. InProceedings of the ACM Symposium on Document Engineering 2023, pp. 1-4, 2023.
[8] Pu Y., Liu D., Chen S., andZhong Y.Research Progress on the Aesthetic Quality Assessment of Complex Layout Images Based on Deep Learning. Applied Sciences, vol. 13, no. 17, pp. 9763, 2023.
[9] Zhong Z., Wang J., Sun H., Hu K., Zhang E., Sun L., andHuo Q.A Hybrid Approach to Document Layout Analysis for Heterogeneous Document Images. In International Conference on Document Analysis and Recognition, Cham: Springer Nature Switzerland, pp. 189-206,2023.
[10] Binmakhashen, G.M. and Mahmoud, S.A.Document Layout Analysis: A Comprehensive Survey. ACM Computing Surveys (CSUR), vol. 52, no. 6, pp. 1-36, 2019.
[11] Deng Q., Ibrayim M., Hamdulla A., andZhang C.The YOLO Model That Still Excels in Document Layout Analysis.Signal, Image and Video Processing, pp. 1-10, 2023.
[12] Oliveira S.A., Seguin B., andKaplan F.DhSegment: A Generic Deep-Learning Approach for Document Segmentation. In2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR), IEEE, pp. 7-12, 2018.
[13] Ghosh R.Newspaper Text Recognition in Bengali Script using Support Vector Machine.Multimedia Tools and Applications, pp. 1-19, 2023.
[14] Shen Z., Zhang R., Dell M., Lee B.C.G., Carlson, J., and Li, W. Layoutparser: A Unified Toolkit for Deep Learning Based Document Image Analysis. In Document Analysis and Recognition-ICDAR 2021: 16th International Conference, Lausanne, Switzerland, September 5-10, 2021, Proceedings, Part I 16, Springer International Publishing, pp. 131-146, 2021.
[15] Lee B.C.G., Mears, J., Jakeway, E., Ferriter, M., Adams, C., Yarasavage, N., Thomas, D., Zwaard, K., and Weld, D.S. The Newspaper Navigator Dataset: Extracting Headlines and Visual Content from 16 Million Historic Newspaper Pages in Chronicling America. InProceedings of the 29th ACM international conference on information & knowledge management, pp. 3055-3062, 2020.
[16] Lombardi, F. and Marinai, S.Deep Learning for Historical Document Analysis and Recognition—A Survey. Journal of Imaging, vol. 6, no. 10, pp. 110, 2020.
[17] Alberti M., Bouillon M., Ingold R., andLiwicki M.Open Evaluation Tool for Layout Analysis of Document Images. In 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), IEEE, vol. 4, pp. 43-47, 2017.
[18] Ravichandra S.,Siva Sathya, S., and Lourdu Marie Sophie, S. Deep Learning Based Document Layout Analysis on Historical Documents. In Advances in Distributed Computing and Machine Learning: Proceedings of ICADCML 2022. Singapore: Springer Nature Singapore, pp. 271-281, 2022.
[19] Cuo, Y., Tashi, N., Liu, Z., Wei, Q., Gadeng, L., and Trashi, G. Layout Analysis of Tibetan Historical Documents Based on Deep Learning. In Proceedings of the 2019 the International Conference on Pattern Recognition and Artificial Intelligence, pp. 96-101, 2019.
[20] Terven, J. and Cordova-Esparza, D. A Comprehensive Review of YOLO: from YOLOv1 to YOLOv8 and Beyond.arXiv preprint arXiv:2304.00501, 2023.