|Communications on Applied Electronics|
|Foundation of Computer Science (FCS), NY, USA|
|Volume 1 - Number 6|
|Year of Publication: 2015|
|Authors: Ajimi Ameer, Sree Kumar.k|
Ajimi Ameer, Sree Kumar.k . Efficient Automatic Image Annotation using Weighted Feature Fusion and its Optimization using Genetic Algorithm. Communications on Applied Electronics. 1, 6 ( April 2015), 15-19. DOI=10.5120/cae-1567
The conventional way of text based retrieval systems are now being replaced by the visual content based systems in image retrieval. The image content has several dominant characteristics features like texture,color,shape and it is interesting to research the classification of images on content-basis with these features. These different descriptor can be combined to form a single feature vector. However, in order to get optimum performance and to reduce the feature dimensionality for making system close to human perception, genetic algorithm (GA) based feature selection is used in this paper. Single feature describes image content only from one point of view, which will not give a correct result. Fusing multi- feature similarity score is expected to improve the accuracy of system's retrieval performance. Inorder to assign the fusion weights of multi-feature similarity scores reasonably according to an image, the genetic algorithm is applied.