Author/Authors :
sabri, nurbaity universiti teknologi mara, kampus jasin - faculty of computer and mathematical sciences, Melaka, Malaysia , yusof, noor hazira universiti teknologi mara - faculty of computer and mathematical sciences, Shah Alam, Malaysia , ibrahim, zaidah universiti teknologi mara - faculty of computer and mathematical sciences, Shah Alam, Malaysia , kasiran, zolidah universiti teknologi mara - faculty of computer and mathematical sciences, Malaysia , abu mangshor, nur nabilah universiti teknologi mara - faculty of computer and mathematical sciences, Malaysia
Abstract :
Text localisation determines the location of the text in an image. This processis performed prior to text recognition. Localising text on shop signage isa challenging task since the images of the shop signage consist of complexbackground, and the text occurs in various font types, sizes, and colours.Two popular texture features that have been applied to localise text inscene images are a histogram of oriented gradient (HOG) and speeded uprobust features (SURF). A comparative study is conducted in this paperto determine which is better with support vector machine (SVM) classifier.The performance of SVM is influenced by its kernel function and anothercomparative study is conducted to identify the best kernel function. Theexperiments have been conducted using primary data collected by theauthors. Results indicate that HOG with quadratic kernel function localisestext for shop signage better than SURF.
