A hybrid codebook model for object categorization using two-way clustering based codebook generation method

Abstract

Both the visual codebook and the codebook model are considered as two main parts of most object classification frameworks. In the original codebook model, each image descriptor is encoded using a single codebook obtained usually using a clustering approach. In this paper, we propose a hybrid codebook model for an object classification task. A simultaneous clustering approach is applied to image descriptors to generate two variant codebooks and used them separately to encode and represent an image through a patch-based codebook model and a feature-based codebook model respectively. The proposed codebook model has been tested on the Caltech-101 dataset. Experimental results demonstrate state-of-the-art performance compared to typical clustering-based codebook model.

Keywords:

• Computer vision
• object categorization
• machine learning
• visual codebook
• simultaneous clustering

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Samira Chebbout

Samira Chebbout received her Engineer in Computer Science in 2005 and Magister in Artificial Intelligence in 2008 from Annaba University. She is currently a lecturer researcher at the Department of Computer Science in Oum El Bouaghi University, Algeria. Her research interests include pattern recognition, machine learning, image processing and computer vision.

Hayet Farida Merouani

Hayet Farida Merouani received her engineering degree from Annaba University, Algeria in 1984, and PhD degree from Robert Gordon University, Aberdeen, UK. Actually, she is a Full Associate Professor at Badji Mokhtar University, Annaba. She also leads a research group of pattern recognition, as a national program research of breast cancer. Her current works focus on the computer vision, medical imaging and biometry.