Unsupervised Segmentation of Text Fragments in Real Scenes
This paper proposes a method that aims to reduce a real scene to a set of regions that contain text fragments and keep small number of false positives. Text is modeled and characterized as a texture pattern, by employing the QMF wavelet decomposition as a texture feature extractor. Processing includes segmentation and spatial selection of regions and then content-based selection of fragments. Unlike many previous works, text fragments in different scales and resolutions laid against complex backgrounds are segmented without supervision. Tested in four image databases, the method is able to reduce visual noise to 4.69% and reaches 96.5% of coherency between the localized fragments and those generated by manual segmentation.
KeywordsManual Segmentation Decomposition Level License Plate Real Scene Visual Noise
- 1.Barroso, J., Rafael, A., Dagless, E.L., Bulas-Cruz, J.: Number plate reading using computer vision. In: IEEE-International Symposium on Industrial Electronics ISIE 1997, pp. 761–766 (1997)Google Scholar
- 2.Setchell, C.J.: Applications of Computer Vision to Road-traffic Monitoring. PhD thesis, Faculty of Engineering, Department of Electrical and Electronic Engineering of the Univesity of Bristol (1997)Google Scholar
- 3.Mariano, V.Y., Kasturi, R.: Detection of text marks on moving vehicles. In: 7th International Conference on Document Analysis and Recognition (ICDAR 2003), 2-Volume Set, August 3-6, Edinburgh, Scotland, UK, pp. 393–397 (2003)Google Scholar
- 4.Clark, P., Mirmehdi, M.: Finding text regions using localised measures. In: Mirmehdi, M., Thomas, B. (eds.) Proceedings of the 11th British Machine Vision Conference, pp. 675–684. BMVA Press (2000)Google Scholar
- 8.Smith, J.R.: Integrated Spatial and Feature Image Systems: Retrieval, Analysis and Compression. Phd. thesis, Graduate School of Arts and Sciences, Columbia University, New York, NY (1997)Google Scholar