ABSTRACT
The problem of text detection in natural scene images is challenging because of the unconstrained sizes, colors, backgrounds and alignments of the characters. This paper proposes novel symmetry features for this task. Within a text line, the intra-character symmetry captures the correspondence between the inner contour and the outer contour of a character while the inter-character symmetry helps to extract information from the gap region between two consecutive characters. A formulation based on Gradient Vector Flow is used to detect both types of symmetry points. These points are then grouped into text lines using the consistency in sizes, colors, and stroke and gap thickness. Therefore, unlike most existing methods which use only character features, our method exploits both the text features and the gap features to improve the detection result. Experimentally, our method compares well to the state-of-the-art on public datasets for natural scenes and street-level images, an emerging category of image data. The proposed technique can be used in a wide range of multimedia applications such as content-based image/video retrieval, mobile visual search and sign translation.
Supplemental Material
Available for Download
The ZIP file contains a PDF file which shows additional full-image results of the proposed text detection method on natural scenes and street images.
- Canny, J. 1986. A Computational Approach To Edge Detection. IEEE TPAMI, 8:679--714, 1986. Google ScholarDigital Library
- Chen, D., Odobez, J.-M., and Bourlard, H. 2004. Text detection and recognition in images and video frames. Pattern Recognition, 37(3), 2004, pp. 595--608.Google ScholarCross Ref
- Chen, H., Tsai, S. S., Schroth, G., Chen, D. M., Grzeszczuk, R., and Girod, B. Robust text detection in natural images with edge-enhanced maximally stable extremal regions. ICIP 2011.Google ScholarCross Ref
- Chen, X., and Yuille, A. L. 2004. Detecting and Reading Text in Natural Scenes. CVPR 2004. Google ScholarDigital Library
- Dalal, N., and Triggs, B. 2005. Histograms of Oriented Gradients for Human Detection. CVPR 2005. Google ScholarDigital Library
- Epshtein, B., Ofek, E., and Wexler, Y. 2010. Detecting Text in Natural Scenes with Stroke Width Transform. CVPR 2010.Google Scholar
- Liang, J., Doermann D., and Li, H. 2005. Camera-based Analysis of Text and Documents: A Survey. IJDAR, 7(2), 2005, pp. 84--104.Google ScholarDigital Library
- Liu, C., Wang, C., and Dai, R. 2005. Text Detection in Images Based on Unsupervised Classification of Edge-based Features. ICDAR 2005. Google ScholarDigital Library
- Lucas, S. M. 2005. ICDAR 2005 Text Locating Competition Results. ICDAR 2005. Google ScholarDigital Library
- Neumann, L., and Matas, J. 2011. Text Localization in Real-World Images Using Efficiently Pruned Exhaustive Search. ICDAR 2011. Google ScholarDigital Library
- Tsai, S. S., Chen, D., Chen, H., Hsu, C.-H., Kim, K.-H., Singh, J. P., and Girod, B. 2011. Combining Image and Text Features: A Hybrid Approach to Mobile Book Spine Recognition. ACM MM 2011. Google ScholarDigital Library
- Wang, K., Babenko, B., and Belongie, S. 2011. End-to-End Scene Text Recognition. ICCV 2011.Google Scholar
- Xu, C., and Prince, J. L. 1998. Snakes, Shapes, and Gradient Vector Flow. IEEE TIP, 7(3), 1998, pp. 359--369. Google ScholarDigital Library
- ICDAR 2003 Dataset. http://algoval.essex.ac.uk/icdar/Datasets.htmlGoogle Scholar
Index Terms
- Detecting text in the real world
Recommendations
A feature-based method of rapidly detecting global exact symmetries in CAD models
Detecting global exact symmetries in CAD models is of great importance in the research of CAD/CAE integration. Therefore, a method is proposed in this paper to rapidly detect the global exact rotational and reflectional symmetries in feature-based CAD ...
Could scene context be beneficial for scene text detection?
Scene text detection and scene segmentation are meaningful tasks in the computer vision field. Could the semantic scene segmentation assist scene text detection in any degree? For example, can we expect the probability of a region being text is low if ...
A Multi-Oriented Scene Text Detector with Position-Sensitive Segmentation
ICMR '18: Proceedings of the 2018 ACM on International Conference on Multimedia RetrievalScene text detection has been studied for a long time and lots of approaches have achieved promising performances. Most approaches regard text as a specific object and utilize the popular frameworks of object detection to detect scene text. However, ...
Comments