MLENN-KELM: a Prototype Selection Based Kernel Extreme Learning Machine Approach for Large-Scale Automatic Image Annotation

With the fast growth of digital images in web, large-scale Automatic Image Annotation (AIA) dealt with some of critical challenges. The most important of them are system scalabilityand annotation performance. On the other hand, learning methods in the large-scale systems with the large number oftraining instances cannot correctly perform and deal with memory and learning time restrictions. In this paper in order to solve the performance of large-scale AIA systems, andlimitations of employing learning methods in these systems, MLENN-KELM approach has been proposed. In the proposedapproach, first, most effective instances are selected from training set by Prototype Selection (PS) methods. The basicassumption of selecting effective instances is reducing the size oftraining set and solving memory restrictions in large-scale AIA systems and learning methods. Then, annotation process is doneby Kernel Extreme Learning Machine algorithm (KELM). The main advantage of using KELM algorithm is the improvement ofannotation performance than other learning methods. Experimental results on NUS-WIDE-Object image set demonstrate the good performance of proposed approach in solving large-scale AIA challenges and also capability improvement of KELM algorithm in large-scale applications.

Keywords:

Automatic Image Annotation , Kernel Extreme Learning Machine , Large-Scale Learning Context , Prototype Selection