KERNEL GBDA FOR RELEVANCE FEEDBACK IN IMAGE RETRIEVAL

2007 ◽  
Vol 07 (04) ◽  
pp. 767-776
Author(s):  
JING LI ◽  
YUAN YUAN
Keyword(s):  
Image Retrieval ◽  
Relevance Feedback ◽  
Large Scale ◽  
Feature Space ◽  
Small Samples ◽  
Generalized Singular Value Decomposition ◽  
Size Problem ◽  
Training Samples ◽  
Feedback Algorithm ◽  
Value Decomposition

Relevance feedback, as a user-in-the-loop mechanism, has been widely employed to improve the performance of content-based image retrieval. Generally, in a relevance feedback algorithm, two key components are: (1) how to select a subset of effective features from a large-scale feature pool and, (2) correspondingly, how to construct a suitable dissimilarity measure. In previous work, the biased discriminant analysis (BDA) has been proposed to address these two problems during the feedback iterations. However, BDA encounters the so called small samples size problem because it has a lack of training samples. In this paper, we utilize the generalized singular value decomposition (GSVD) to significantly reduce the small samples size problem in BDA. The developed algorithm is named GSVD for BDA (GBDA). We then kernelize the GBDA to nonlinear kernel feature space. A large amount of experiments were carried out upon a large scale database, which contains 17800 images. From the experimental results, GBDA and its kernelization are demonstrated to outperform the traditional BDA-based relevance feedback approaches and their kernel extensions, respectively.

scholarly journals Improving the Results of Google Scholar Engine through Automatic Query Expansion Mechanism and Pseudo Re-ranking using MVRA

2018 ◽  
Vol 42 (2) ◽  
pp. 219-229
Author(s):  
Mawloud Mosbah
Keyword(s):  
Image Retrieval ◽  
Relevance Feedback ◽  
Query Expansion ◽  
Majority Voting ◽  
Google Scholar ◽  
Content Based Image Retrieval ◽  
Ranking Algorithm ◽  
Expansion Mechanism ◽  
Feedback Algorithm ◽  
Pseudo Relevance Feedback

In this paper, we address the enhancing of Google Scholar engine, in the context of text retrieval, through two mechanisms related to the interrogation protocol of that query expansion and reformulation. The both schemes are applied with re-ranking results using a pseudo relevance feedback algorithm that we have proposed previously in the context of Content based Image Retrieval (CBIR) namely Majority Voting Re-ranking Algorithm (MVRA). The experiments conducted using ten queries reveal very promising results in terms of effectiveness.

A Low Memory Cost Fast Retrieval Method Based on Bucket Map Chain

2013 ◽  
Vol 321-324 ◽  
pp. 969-973
Author(s):  
Tian Qiang Peng ◽  
Xiao Feng Sun
Keyword(s):  
Information Retrieval ◽  
Image Retrieval ◽  
Large Scale ◽  
Feature Space ◽  
Locality Sensitive Hashing ◽  
Query Point ◽  
Retrieval Method ◽  
Memory Cost

To reduce the high memory cost of fast retireval method, we present a fast retrieval method based on bucket map chain on the basis of Exact Euclidean Locality Sensitive Hashing (E2LSH). The bucket map chain contains all the points projected from feature space in multiple buckets, which store the nearby points. When conducting query, it searches the chain by the bucket index of query point and locates the position of related buckets, then reads the related points in related buckets and measurs the similarity of these points with query point. The experiments show that this method can efficiently decrease the memory cost of retrieval. It is very important for increasing the feasibility of large scale information retrieval especially image retrieval.

A CONTEXT-BASED APPROACH FOR COLOR IMAGE RETRIEVAL

2002 ◽  
Vol 16 (02) ◽  
pp. 239-255 ◽  
Author(s):  
JAU-LING SHIH ◽  
LING-HWEI CHEN
Keyword(s):  
Image Retrieval ◽  
Similarity Measure ◽  
Relevance Feedback ◽  
Color Image ◽  
Experimental Results ◽  
Retrieval Method ◽  
Feature Vectors ◽  
The Mean ◽  
Feedback Algorithm

In this paper, a color image retrieval method based on the primitives of images will be proposed. First, the context of each pixel in an image will be defined. Then, the contexts in the image are clustered into several classes based on the algorithm of fast noniterative clustering. The mean of the context in the same class is considered as a primitive of the image. The primitives are used as feature vectors. Since the numbers of primitives between images are different, a specially designed similarity measure is then proposed to do color image retrieval. To better adapt to the preferences of users, a relevance feedback algorithm is provided to automatically determine the weight of each primitive according to the user's response. To demonstrate the effectiveness of the proposed system, several test databases from Corel are used to compare the performances of the proposed system with other methods. The experimental results show that the proposed system is superior to others.

EXTRACTING SALIENT REGIONS AND LEARNING IMPORTANCE SCORES IN REGION-BASED IMAGE RETRIEVAL

2003 ◽  
Vol 17 (08) ◽  
pp. 1349-1367 ◽  
Author(s):  
BYOUNGCHUL KO ◽  
HYERAN BYUN
Keyword(s):  
Image Retrieval ◽  
Relevance Feedback ◽  
Retrieval System ◽  
Retrieval Performance ◽  
Matching Process ◽  
Similar Images ◽  
Salient Regions ◽  
Feedback Method ◽  
Successful Retrieval ◽  
Feedback Algorithm

In this paper, we propose a new method for extracting salient regions and learning their importance scores in region-based image retrieval. In Region-Based Image Retrieval (RBIR), not all the regions are important for retrieving similar images and rather, in retrieval, the user is often interested in performing a query on only one or a few regions rather than the whole image. Therefore, for a successful retrieval system, it is an important issue to specify which regions are important for retrieving an image. To extract salient regions from images automatically, we make three assumptions and determine salient regions with their importance scores. In this paper, we apply the relevance feedback algorithm to the matching process as two different purposes: one is for updating importance scores of salient regions and the other is for updating weights of feature vectors. By using our relevance feedback method, the matching process can improve retrieval performance interactively and allow progressive refinement of query results according to the user's feedback action. Through experiments and comparison with other methods, our proposed method shows good performance as well as easy and semantic interface for region-based image retrieval. The efficacy of our method is validated using a set of 3000 images from Corel-photo CD.

Sign in / Sign up

Export Citation Format

Share Document