Text detection and recognition in urban scenes
Authors:
- 1 November 2011
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 227-234
Abstract
Text detection and recognition in real images taken in unconstrained environments, such as street view images, remain surprisingly challenging in Computer Vision. In this paper, we present a comprehensive strategy combining bottom-up and top-down mechanisms to detect Text boxes. The bottom-up part is based on character segmentation and grouping . The top-down part is achieved with a statistical learning approach based on box descriptors. Our main contribution consists in introducing a new descriptor, Fuzzy HOG (F-HOG), fully adapted for text box analysis. A thorough experimental validation proves the efficiency of the whole system outperforming state of the art results on the standard ICDAR text detection benchmark. Another contribution concerns the exploitation of our text extraction in a complete search engine scheme. We propose to retrieve a location from a textual query: combining our text box detection technology with OCR on georeferenced street images, we achieved a GIS system with a fully automatic textual indexing. We demonstrate the relevance of our system on the real urban database of [10].Keywords
This publication has 9 references indexed in Scilit:
- Robust text detection in natural images with edge-enhanced Maximally Stable Extremal Regions2011
- Text Detection Using Edge Gradient and Graph Spectrum2010
- Detecting text in natural scenes with stroke width transform2010
- A New Block Partitioned Text Feature for Text Verification2009
- Real-time Accurate Object Detection using Multiple Resolutions2007
- Histograms of Oriented Gradients for Human Detection2005
- ICDAR 2005 text locating competition results2005
- Support-vector networksMachine Learning, 1995
- Postal address block location in real timeComputer, 1992