Autors: Ismailov, A. V., Hristov, V. D.
Title: Brochure Segmentation Methodology
Keywords: brochure, image, segmentation, YOLO models

Abstract: This paper research the problem of hypermarket brochure segmentation with the aim of researching and selecting an appropriate brochure segmentation model. The entire scientific methodology is to be used to create a dataset containing brochures from well-known supermarkets in Bulgaria. The main task in the scientific publication is to enable marking of areas containing product price information on the brochure page - product image, product name, product price and additional product information, if any. The studies are selected parameters for 50, 100 and 200 epochs for the selected models pre-tuned according to the specified formula, and in the latter deficit, the YOLOv8-N model is as the best selected because it consists of a small number and has the best accuracy.

References

  1. D. Wang, "MCLSS: Multi-level Contrastive Learning for Semantic Segmentation, " 2023 IEEE International Conference on Sensors, Electronics and Computer Engineering (ICSECE), Jinzhou, China, 2023, pp. 419-422, doi: 10.1109/ICSECE58870.2023.10263424.
  2. LC Chen, G Papandreou, I Kokkinos et al., "Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs[J]", Computer Science, no. 4, pp. 357-361, 2014.
  3. B. U. Mahmud and G. Y. Hong, "Semantic Image Segmentation using CNN (Convolutional Neural Network) based Technique, " 2022 IEEE World Conference on Applied Intelligence and Computing (AIC), Sonbhadra, India, 2022, pp. 210-214, doi: 10.1109/AIC55036.2022.9848977.
  4. B. Baheti, S. Innani, S. Gajre and S. Talbar, "Eff-UNet: A Novel Architecture for Semantic Segmentation in Unstructured Environment", 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 1473-1481, 2020.
  5. A. Ismailov, V. Hristov, Products-price in brochures dataset, Available: Https://app. roboflow.com/bulgarian-brochures/products-price-inbrochures
  6. G. Jocher, Ultralytics YOLOv5. 2020. doi: 10.5281/zenodo.3908559.
  7. C. Li et al., YOLOv6 v3.0: A Full-Scale Reloading. 2023.
  8. G. Jocher, A. Chaurasia, and J. Qiu, Ultralytics YOLOv8. 2023. [Online]. Available: Https://github.com/ultralytics/ultralytics

Issue

2025 14th Mediterranean Conference on Embedded Computing, MECO 2025 - Proceedings, 2025, Macedonia, https://doi.org/10.1109/MECO66322.2025.11049297

Вид: публикация в международен форум, публикация в реферирано издание, индексирана в Scopus и Web of Science