A novel method for image feature fusion based on MADM

Single type of feature can hardly describe the abundant image information completely. Any kind of image features being used in image retrieval has its advantages and disadvantage. Different features can be combined with others in order to improve the retrieval effectiveness. This paper proposed a novel method which is dynamic, easy to extend and operate simply. This method is based on MADM (Multiple Attribute Decision Making), uses subjective and objective weighting method and uses TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) to generate the final result set. This method fuses features at the time of the retrieval and use parallelization to reduce the cost. Experimental results show that this method can fuse features effectively and can improve the retrieval effect.