Comparing fusion techniques for the ImageCLEF 2013 medical case retrieval task

A Novel Adaptive Feature Fusion Strategy for Image Retrieval

In the era of big data, it is challenging to efficiently retrieve the required images from the vast amount of data. Therefore, a content-based image retrieval system is an important research direction to address this problem. Furthermore, a multi-feature-based image retrieval system can compensate for the shortage of a single feature to a certain extent, which is essential for improving retrieval system performance. Feature selection and feature fusion strategies are critical in the study of multi-feature fusion image retrieval. This paper proposes a multi-feature fusion image retrieval strategy with adaptive features based on information entropy theory. Firstly, we extract the image features, construct the distance function to calculate the similarity using the information entropy proposed in this paper, and obtain the initial retrieval results. Then, we obtain the precision of single feature retrieval based on the correlation feedback as the retrieval trust and use the retrieval trust to select the effective features automatically. After that, we initialize the weights of selected features using the average weights, construct the probability transfer matrix, and use the PageRank algorithm to update the initialized feature weights to obtain the final weights. Finally, we calculate the comprehensive similarity based on the final weights and output the detection results. This has two advantages: (1) the proposed strategy uses multiple features for image retrieval, which has better performance and more substantial generalization than the retrieval strategy based on a single feature; (2) compared with the fixed-feature retrieval strategy, our method selects the best features for fusion in each query, which takes full advantages of each feature. The experimental results show that our proposed method outperforms other methods. In the datasets of Corel1k, UC Merced Land-Use, and RSSCN7, the top10 retrieval precision is 99.55%, 88.02%, and 88.28%, respectively. In the Holidays dataset, the mean average precision (mAP) was 92.46%.

An Adaptive Weight Method for Image Retrieval Based Multi-Feature Fusion

With the rapid development of information storage technology and the spread of the Internet, large capacity image databases that contain different contents in the images are generated. It becomes imperative to establish an automatic and efficient image retrieval system. This paper proposes a novel adaptive weighting method based on entropy theory and relevance feedback. Firstly, we obtain single feature trust by relevance feedback (supervised) or entropy (unsupervised). Then, we construct a transfer matrix based on trust. Finally, based on the transfer matrix, we get the weight of single feature through several iterations. It has three outstanding advantages: (1) The retrieval system combines the performance of multiple features and has better retrieval accuracy and generalization ability than single feature retrieval system; (2) In each query, the weight of a single feature is updated dynamically with the query image, which makes the retrieval system make full use of the performance of several single features; (3) The method can be applied in two cases: supervised and unsupervised. The experimental results show that our method significantly outperforms the previous approaches. The top 20 retrieval accuracy is 97.09%, 92.85%, and 94.42% on the dataset of Wang, UC Merced Land Use, and RSSCN7, respectively. The Mean Average Precision is 88.45% on the dataset of Holidays.