Untitled

Markkula, Marjo; Tico, Marius; Sepponen, Bemmu; Nirkkonen, Katja; Sormunen, Eero

doi:10.1023/a:1011954407169

Cited by 17 publications

(1 citation statement)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the context of CBIR systems, texture plays an important role (Wu et al 1992, Markkula et al 2001, as many image samples frequently exhibit a set of non uniform features. This diversity poses a hurdle for many image retrieval processes when other attributes such as color and shape are employed.…”

Section: Introductionmentioning

confidence: 99%

Medical image retrieval and analysis by Markov random fields and multi-scale fractal dimension

et al. 2015

View full text Add to dashboard Cite

Many Content-based Image Retrieval (CBIR) systems and image analysis tools employ color, shape and texture (in a combined fashion or not) as attributes, or signatures, to retrieve images from databases or to perform image analysis in general. Among these attributes, texture has turned out to be the most relevant, as it allows the identification of a larger number of images of a different nature. This paper introduces a novel signature which can be used for image analysis and retrieval. It combines texture with complexity extracted from objects within the images. The approach consists of a texture segmentation step, modeled as a Markov Random Field process, followed by the estimation of the complexity of each computed region. The complexity is given by a Multi-scale Fractal Dimension. Experiments have been conducted using an MRI database in both pattern recognition and image retrieval contexts. The results show the accuracy of the proposed method in comparison with other traditional texture descriptors and also indicate how the performance changes as the level of complexity is altered.

show abstract

Section: Introductionmentioning

confidence: 99%

Medical image retrieval and analysis by Markov random fields and multi-scale fractal dimension

et al. 2015

View full text Add to dashboard Cite

show abstract

Task‐based information retrieval: Structuring undergraduate history essays for better course evaluation using essay‐type visualizations

Leide

Cole

Beheshti

et al. 2007

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

Stage 3, the exploration stage of researching an assignment, they often do not know their information need; this causes them to go back to Stage 2, the topic-selection stage, when they are selecting keywords to formulate their query to an Information Retrieval (IR) system. Our hypothesis is that instead of going backward, they should be going forward toward a goal state-the performance of the task for which they are seeking the information. If they can somehow construct their goal state into a query, this forward-looking query better operationalizes their information need than does a topic-based query. For domain novice undergraduates seeking information for a course essay, we define their task as selecting a high-impact essay structure which will put the students' learning on display for the course instructor who will evaluate the essay. We report a study of firstyear history undergraduate students which tested the use and effectiveness of "essay type" as a task-focused query-formulation device. We randomly assigned 78 history undergraduates to an intervention group and a control group. The dependent variable was essay quality, based on (a) an evaluation of the student's essay by a research team member, and (b) the marks given to the student's essay by the course instructor. We found that conscious or formal consideration of essay type is inconclusive as a basis of a task-focused query-formulation device for IR.

show abstract

Factors affecting rocchio‐based pseudorelevance feedback in image retrieval

Tsai

Chen

2014

Asso for Info Science & Tech

View full text Add to dashboard Cite

Pseudorelevance feedback (PRF) was proposed to solve the limitation of relevance feedback (RF), which is based on the user-in-the-loop process. In PRF, the top-k retrieved images are regarded as PRF. Although the PRF set contains noise, PRF has proven effective for automatically improving the overall retrieval result. To implement PRF, the Rocchio algorithm has been considered as a reasonable and well-established baseline. However, the performance of Rocchio-based PRF is subject to various representation choices (or factors). In this article, we examine these factors that affect the performance of Rocchio-based PRF, including image-feature representation, the number of top-ranked images, the weighting parameters of Rocchio, and similarity measure. We offer practical insights on how to optimize the performance of Rocchio-based PRF by choosing appropriate representation choices. Our extensive experiments on NUS-WIDE-LITE and Caltech 101 + Corel 5000 data sets show that the optimal feature representation is color moment + wavelet texture in terms of retrieval efficiency and effectiveness. Other representation choices are that using top-20 ranked images as pseudopositive and pseudonegative feedback sets with the equal weight (i.e., 0.5) by the correlation and cosine distance functions can produce the optimal retrieval result.

show abstract

Untitled

Cited by 17 publications

References 24 publications

Medical image retrieval and analysis by Markov random fields and multi-scale fractal dimension

Medical image retrieval and analysis by Markov random fields and multi-scale fractal dimension

Task‐based information retrieval: Structuring undergraduate history essays for better course evaluation using essay‐type visualizations

Factors affecting rocchio‐based pseudorelevance feedback in image retrieval

Contact Info

Product

Resources

About