Extraction of sparse features of color images in recognizing objects

• Published in: International journal of control, automation, and systems Vol. 14; no. 2; pp. 616 - 627
• Main Authors: Quyen Bui, T. T., Vu, Thang T., Hong, Keum-Shik
• Format: Journal Article
• Language: English
• Published: Bucheon / Seoul Institute of Control, Robotics and Systems and The Korean Institute of Electrical Engineers 01.04.2016; Springer Nature B.V; 제어·로봇·시스템학회
• Subjects: Analysis; Automation; Color; Color imagery; Control; Cortexes; Datasets; Energy; Engineering; Extraction; Mechatronics; Object recognition; Orientation; Pattern recognition systems; Prototypes; Regular Papers; Robotics; Studies; Support vector machines; 제어계측공학; object recognition; Gabor energy filter; color-based features; CIELAB color space; local maximum pooling; sparse features
• ISSN: 1598-6446; 2005-4092
• DOI: 10.1007/s12555-014-0502-9

In this paper, we propose a new object recognition framework that combines Gabor energy filters, a visual cortex model in which single units alternate with complex units, and color information. Each color image is first converted to the CIELAB color space. Rather using Gabor filters in the first layer of the cortex model, to each color component, a set of Gabor energy filters is applied. Thereafter, the superposition responses of the Gabor energy filter outputs over the color components are normalized by divisive normalization. In the fourth layer, sparse features are calculated using a localized pooling method that allows retention of some geometric information from the prototype patches’ positions. Finally, a set of sparse features are exploited by a linear SVM classifier for object recognition and classification. In the learning stage, a set of prototypes is selected randomly over spatial position, spatial size, and several scales simultaneously, and is extracted by the local maximum over scales and orientations, ignoring weaker training scales and orientations. The results of experiments performed on several datasets show that the use of color information in our framework improves object recognition significantly.