基于虚拟关联反馈与聚类的图像检索技术研究

基于虚拟相关反馈(PRF)技术,提出了一种新的自动关联反馈检索方法--外部自动相关反馈(OARF).该方法基于图像内容特征距离,应用K-均值聚类,自动扩展查询图像特征,从而提高检索性能.试验结果表明,OARF能够降低用户负担,显著提高原始检索算法的性能,缩小"语义鸿沟".     

Collaborative pseudo-relevance feedback

Pseudo-relevance feedback (PRF) is a technique commonly used in the field of information retrieval. The performance of PRF is heavily dependent upon parameter values. When relevance judgements are unavailable, these parameters are difficult to set. In the following paper, we introduce a novel approach to PRF inspired by collaborative filtering (CF). We also describe an adaptive tuning method which automatically sets algorithmic parameters. In a multi-stage evaluation using publicly available datasets, our technique consistently outperforms conventional PRF, regardless of the underlying retrieval model.     

Improving retrievability with improved cluster-based pseudo-relevance feedback selection

High findability of documents within a certain cut-off rank is considered an important factor in recall-oriented application domains such as patent or legal document retrieval. Findability is hindered by two aspects, namely the inherent bias favoring some types of documents over others introduced by the retrieval model, and the failure to correctly capture and interpret the context of conventionally rather short queries. In this paper, we analyze the bias impact of different retrieval models and query expansion strategies. We furthermore propose a novel query expansion strategy based on document clustering to identify dominant relevant documents. This helps to overcome limitations of conventional query expansion strategies that suffer strongly from the noise introduced by imperfect initial query results for pseudo-relevance feedback documents selection. Experiments with different collections of patent documents suggest that clustering based document selection for pseudo-relevance feedback is an effective approach for increasing the findability of individual documents and decreasing the bias of a retrieval system.     

An efficient concept-based retrieval model for enhancing text retrieval quality

Most of the common techniques in text retrieval are based on the statistical analysis terms (words or phrases). Statistical analysis of term frequency captures the importance of the term within a document only. Thus, to achieve a more accurate analysis, the underlying model should indicate terms that capture the semantics of text. In this case, the model can capture terms that represent the concepts of the sentence, which leads to discovering the topic of the document. In this paper, a new concept-based retrieval model is introduced. The proposed concept-based retrieval model consists of conceptual ontological graph (COG) representation and concept-based weighting scheme. The COG representation captures the semantic structure of each term within a sentence. Then, all the terms are placed in the COG representation according to their contribution to the meaning of the sentence. The concept-based weighting analyzes terms at the sentence and document levels. This is different from the classical approach of analyzing terms at the document level only. The weighted terms are then ranked, and the top concepts are used to build a concept-based document index for text retrieval. The concept-based retrieval model can effectively discriminate between unimportant terms with respect to sentence semantics and terms which represent the concepts that capture the sentence meaning. Experiments using the proposed concept-based retrieval model on different data sets in text retrieval are conducted. The experiments provide comparison between traditional approaches and the concept-based retrieval model obtained by the combined approach of the conceptual ontological graph and the concept-based weighting scheme. The evaluation of results is performed using three quality measures, the preference measure (bpref), precision at 10 documents retrieved (P(10)) and the mean uninterpolated average precision (MAP). All of these quality measures are improved when the newly developed concept-based retrieval model is used, confirming that such model enhances the quality of text retrieval.     

Fuzzy conceptual indexing for concept-based cross-lingual text retrieval

Cross-lingual text retrieval (CLTR) is a technique for locating relevant documents in different languages. The authors have developed fuzzy conceptual indexing (FCI) to extend CLTR to include documents that share concepts but don't contain exact translations of query terms. In FCI, documents and queries are represented as a function of language-independent concepts, thus enabling direct mapping between them across multiple languages. Experimental results suggest that concept-based CLTR outperforms translation-based CLTR in identifying conceptually relevant documents.     

Enabling concept-based relevance feedback for information retrievalon the WWW

The World Wide Web is a world of great richness, but finding information on the Web is also a great challenge. Keyword-based querying has been an immediate and efficient way to specify and retrieve related information that the user inquires. However, conventional document ranking based on an automatic assessment of document relevance to the query may not be the best approach when little information is given, as in most cases. In order to clarify the ambiguity of the short queries given by users, we propose the idea of concept-based relevance feedback for Web information retrieval. The idea is to have users give two to three times more feedback in the same amount of time that would be required to give feedback for conventional feedback mechanisms. Under this design principle, we apply clustering techniques to the initial search results to provide concept-based browsing. We show the performance of various feedback interface designs and compare their pros and cons. We measure precision and relative recall to show how clustering improves performance over conventional similarity ranking and, most importantly, we show how the assistance of concept-based presentation reduces browsing labor     

Simulating the future of concept-based video retrieval under improved detector performance

In this paper we address the following important questions for concept-based video retrieval: (1) What is the impact of detector performance on the performance of concept-based retrieval engines, and (2) will these engines be applicable to real-life search tasks if detector performance improves in the future? We use Monte Carlo simulations to answer these questions. To generate the simulation input, we propose to use a probabilistic model of two Gaussians for the confidence scores that concept detectors emit. Modifying the model??s parameters affects the detector performance and the search performance. We study the relation between these two performances on two video collections. For detectors with similar discriminative power and a concept vocabulary of around 100 concepts, the simulation reveals that in order to achieve a search performance of 0.20 mean average precision (MAP)??which is considered sufficient performance for real-life applications??one needs detectors with at least 0.60 MAP . We also find that, given our simulation model and low detector performance, MAP is not always a good evaluation measure for concept detectors since it is not strongly correlated with the search performance.     

Multireference object pose retrieval with volume feedback

Pose retrieval of a rigid object from monocular video sequences or images is addressed. Initially, the object pose is estimated in each image assuming flat depth maps. Shape-from-silhouette is then applied to make a 3-D model (volume), which is used for a new round of pose estimations, this time by a model-based method that gives better estimates. Before repeating this process by building a new volume, pose estimates are adjusted to reduce error by maximizing a novel quality factor for shape-from-silhouette volume reconstruction. The feedback loop is terminated when pose estimates do not change much, as compared with those produced by the previous iteration. Based on a theoretical study of the proposed system, a test of convergence to a given set of poses is devised. Reliable performance of the system is also proved by several experiments on both synthetic and real image sequences. No model is assumed for the object and no feature point is detected or tracked as there is no problematic feature matching or correspondence. Our method can be used for 3-D object tracking in video, 3-D modeling, and volume reconstruction from video.     

Bibliographic retrieval: a survey of individual users of MEDLINE

The National Library of Medicine (NLM) has been providing online access to the MEDLINE database for nearly 20 years. In recent years, there has been a shift in the composition of the user population. Nearly half the online access codes are now held by individuals who conduct their own searches of the database. The NLM has conducted a survey to identify the demographic features of this end-user population, their reasons for searching the database, their methods of access, and their satisfaction with MEDLINE as available on the NLM system.     

Content-based image retrieval with relevance feedback using random walks

In this paper, we propose a novel approach to content-based image retrieval with relevance feedback, which is based on the random walker algorithm introduced in the context of interactive image segmentation. The idea is to treat the relevant and non-relevant images labeled by the user at every feedback round as “seed” nodes for the random walker problem. The ranking score for each unlabeled image is computed as the probability that a random walker starting from that image will reach a relevant seed before encountering a non-relevant one. Our method is easy to implement, parameter-free and scales well to large datasets. Extensive experiments on different real datasets with several image similarity measures show the superiority of our method over different recent approaches.     

Multimodal retrieval with relevance feedback based on genetic programming

This paper presents a framework for multimodal retrieval with relevance feedback based on genetic programming. In this supervised learning-to-rank framework, genetic programming is used for the discovery of effective combination functions of (multimodal) similarity measures using the information obtained throughout the user relevance feedback iterations. With these new functions, several similarity measures, including those extracted from different modalities (e.g., text, and content), are combined into one single measure that properly encodes the user preferences. This framework was instantiated for multimodal image retrieval using visual and textual features and was validated using two image collections, one from the Washington University and another from the ImageCLEF Photographic Retrieval Task. For this image retrieval instance several multimodal relevance feedback techniques were implemented and evaluated. The proposed approach has produced statistically significant better results for multimodal retrieval over single modality approaches and superior effectiveness when compared to the best submissions of the ImageCLEF Photographic Retrieval Task 2008.     

Learning similarity measure for natural image retrieval with relevance feedback

A new scheme of learning similarity measure is proposed for content-based image retrieval (CBIR). It learns a boundary that separates the images in the database into two clusters. Images inside the boundary are ranked by their Euclidean distances to the query. The scheme is called constrained similarity measure (CSM), which not only takes into consideration the perceptual similarity between images, but also significantly improves the retrieval performance of the Euclidean distance measure. Two techniques, support vector machine (SVM) and AdaBoost from machine learning, are utilized to learn the boundary. They are compared to see their differences in boundary learning. The positive and negative examples used to learn the boundary are provided by the user with relevance feedback. The CSM metric is evaluated in a large database of 10009 natural images with an accurate ground truth. Experimental results demonstrate the usefulness and effectiveness of the proposed similarity measure for image retrieval.     

Hidden annotation for image retrieval with long-term relevance feedback learning

Hidden annotation (HA) is an important research issue in content-based image retrieval (CBIR). We propose to incorporate long-term relevance feedback (LRF) with HA to increase both efficiency and retrieval accuracy of CBIR systems. The work contains two parts. (1) Through LRF, a multi-layer semantic representation is built to automatically extract hidden semantic concepts underlying images. HA with these concepts alleviates the burden of manual annotation and avoids the ambiguity problem of keyword-based annotation. (2) For each learned concept, semi-supervised learning is incorporated to automatically select a small number of candidate images for annotators to annotate, which improves efficiency of HA.     

基于支持向量的图像相关反馈检索

对图像相关反馈检索过程建立二分类的支持向量机问题模型,进而提出基于支持向量的图像相关反馈检索方法.比较了基于支持向量的反馈检索方法和传统的反馈检索方法的检索性能,研究了特征提取对支持向量机性能的影响.实验结果表明,基于支持向量的图像相关反馈检索方法具有较好的图像检索效果.     

Extracting Clinical Trial Design Information from MEDLINE Abstracts

Evidence-based medicine (EBM) requires medical practitioners to select appropriate treatments for individual patients based on the current best evidence, and the results of phase III clinical trials are the major source of such evidence. In this paper, we report results of experiment in extracting important information for EBM from the abstracts of phase III clinical trials, in an effort to investigate how far the existing natural language processing (NLP) techniques could support EBM using MEDLINE database.     

Content-based image retrieval using fuzzy perceptual feedback

In this paper, a new framework called fuzzy relevance feedback in interactive content-based image retrieval (CBIR) systems is introduced. Conventional binary labeling scheme in relevance feedback requires a crisp decision to be made on the relevance of the retrieved images. However, it is inflexible as user interpretation of visual content varies with respect to different information needs and perceptual subjectivity. In addition, users tend to learn from the retrieval results to further refine their information requests. It is, therefore, inadequate to describe the user’s fuzzy perception of image similarity with crisp logic. In view of this, we propose a fuzzy relevance feedback approach which enables the user to make a fuzzy judgement. It integrates the user’s fuzzy interpretation of visual content into the notion of relevance feedback. An efficient learning approach is proposed using a fuzzy radial basis function (FRBF) network. The network is constructed based on the user’s feedbacks. The underlying network parameters are optimized by adopting a gradient-descent training strategy due to its computational efficiency. Experimental results using a database of 10,000 images demonstrate the effectiveness of the proposed method.

Adaptive relevance feedback for large-scale image retrieval

Content-based image retrieval aims at substituting traditional indexing based on manual annotation by using automatically-extracted visual indexing features. Novel techniques are needed however to efficiently deal with the semantic gap (i.e. the partial match between the low-level features and the visual content). Here, we investigate a query-free retrieval approach first proposed by Ferecatu and Geman. This approach relies solely on an iterative relevance feedback mechanism that drives a heuristic sampling of the collection, and aims to take explicitly into account the semantic gap. Our contributions are related to three complementary aspects. First, we formalize a large-scale approach based on a hierarchical tree-like organization of the images computed off-line. Second, we propose a versatile modulation of the exploration/exploitation trade-off based on the consistency of the system internal states between successive iterations. Third, we elaborate a long-term optimization of the similarity metric based on the user searching session logs accumulated off-line. We implemented a web-application that integrates all our contributions, and distribute it under the AGPL Version 3 free software license. We organized user-based evaluation campaigns using ImageNet dataset, and show empirically that our contributions significantly improve the retrieval performance of the original framework, that they are complementary to each other, and that their overall integration is consistently beneficial.     

Relevance feedback for real-world human action retrieval

Content-based video retrieval is an increasingly popular research field, in large part due to the quickly growing catalogue of multimedia data to be found online. Even though a large portion of this data concerns humans, however, retrieval of human actions has received relatively little attention. Presented in this paper is a video retrieval system that can be used to perform a content-based query on a large database of videos very efficiently. Furthermore, it is shown that by using ABRS-SVM, a technique for incorporating Relevance feedback (RF) on the search results, it is possible to quickly achieve useful results even when dealing with very complex human action queries, such as in Hollywood movies.     

Bayesian relevance feedback for content-based image retrieval

Despite the efforts to reduce the so-called semantic gap between the user's perception of image similarity and the feature-based representation of images, the interaction with the user remains fundamental to improve performances of content-based image retrieval systems. To this end, relevance feedback mechanisms are adopted to refine image-based queries by asking users to mark the set of images retrieved in a neighbourhood of the query as being relevant or not. In this paper, the Bayesian decision theory is used to estimate the boundary between relevant and non-relevant images. Then, a new query is computed whose neighbourhood is likely to fall in a region of the feature space containing relevant images. The performances of the proposed query shifting method have been compared with those of other relevance feedback mechanisms described in the literature. Reported results show the superiority of the proposed method.     

Information retrieval models in the context of retrieval tasks

Information retrieval models are reviewed from the viewpoint of retrieval needs that cause different types of retrieval tasks. A generalized iterative query-response scheme of the retrieval process is presented. The characteristics of the system of retrieval mechanism models aimed at the support of retrieval tasks of different types, as well as at the development of the retrieval process using internal and external feedback, are stated. The use of models of multidimensional quantitative analysis based on coordinate indexing to perform external feedback is proposed.