Title :
Fast logo detection based on morphological features in document images
Author :
Hassanzadeh, Sina ; Pourghassem, Hossein
Author_Institution :
Dept. of Electr. Eng., Islamic Azad Univ. Najafabad Branch, Isfahan, Iran
Abstract :
In this paper, a novel fast logo detection approach in document images is presented. Logos with separated parts usually can affect the logo detection process. To overcome this problem, some specifications of logos are considered. Our proposed method divided in three main sections. In the first section, a horizontal dilation operator is used to merge separated parts of logo in horizontal direction. In the second section, a simple decision classifier is applied for classifying logo and non-logo. In the final section, a modifying operation for detecting separated-part-logo, logo which has separated part, based on two specifications is used. These specifications include centroid coordinate and intersection of each logo´s separated part bounding box. The proposed approach is evaluated on a public document image database and international logos. Experimental results show its performance in logo detection problem.
Keywords :
decision trees; document image processing; image classification; object detection; centroid coordinate; decision tree classifier; horizontal dilation operator; international logo; logo classification; morphological feature; public document image database; separated-part-logo detection; Classification algorithms; Decision trees; Feature extraction; Pixel; Signal processing; Text analysis; Logo detection; decision tree classifier; feature extraction; horizontal dilation; spatial density;
Conference_Titel :
Signal Processing and its Applications (CSPA), 2011 IEEE 7th International Colloquium on
Conference_Location :
Penang
Print_ISBN :
978-1-61284-414-5
DOI :
10.1109/CSPA.2011.5759888
