Loading...

From windows to logos: analyzing outdoor images to aid flyer classification

Pourashraf, P ; Sharif University of Technology | 2018

514 Viewed
  1. Type of Document: Article
  2. DOI: 10.1007/978-3-319-93000-8_20
  3. Publisher: Springer Verlag , 2018
  4. Abstract:
  5. The goal of this paper was to create a new method for analyzing the online real estate flyers based on their property types. We created an algorithm which identifies the buildings and windows from the buildings in order to extract some useful features for classifying the flyers. Our novel approach for building recognition has two main steps: 1- Building Detector 2- Region Growing. Our novel window detection algorithm uses vanishing point to identify nearly the best angle for applying window detection. It transforms the 2D image into 3D and rotates the 3D image around the z-axis and picks the appropriate angle based on the vanishing points. Using these two novel techniques we were be able to extract a new feature vector which is used to build our final model. This final model is able to classify Retail spaces very well based on the Window and logo features. © 2018, Springer International Publishing AG, part of Springer Nature
  6. Keywords:
  7. Buildings ; Image classification ; Object detection ; Building recognition ; Feature vectors ; Flyer ; Novel techniques ; Outdoor images ; Region growing ; Vanishing point ; Window detection ; Image analysis
  8. Source: 15th International Conference on Image Analysis and Recognition, ICIAR 2018, 27 June 2018 through 29 June 2018 ; Volume 10882 LNCS , 2018 , Pages 175-184 ; 03029743 (ISSN); 9783319929996 (ISBN)
  9. URL: https://link.springer.com/chapter/10.1007%2F978-3-319-93000-8_20