MSIDX: Multi-Sort Indexing for Efficient Content-Based Image Search and Retrieval

Tiakas, Eleftherios; Rafailidis, Dimitrios; Dimou, Anastasios; Daras, Petros

doi:10.1109/tmm.2013.2247989

Cited by 27 publications

(40 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The basic idea of similarity search based on Dimensions Value Cardinalities (DVC) according to [11] is: to reorder the storage positions of images' descriptors according to value cardinalities of their dimensions, by performing a multiple sort algorithm, to increase the probability of having two similar images in storage positions that do not differ more than a specific global constant range, denoted by a parameter 2W . Dimensions Value Cardinalities (DVC) are defined as the unique numbers that occur in the dimensions of the image descriptor vectors.…”

Section: Similarity Search Based On Dvcmentioning

confidence: 99%

“…Real values: In case that the extraction process of the descriptor generates real values, the calculation strategy of the value cardinality cj is performed after limiting the decimal accuracy of the descriptor values, as in the case of integer values. However, in practice, the extracted descriptors have a limited decimal accuracy, usually between 4 and 6 decimals, due to space and computational restrictions [11]. In our experiments (Section 5) no additional value quantization was used in all evaluation datasets.…”

Section: Similarity Search Based On Dvcmentioning

confidence: 99%

“…For instance, high dimensional descriptors that come from the bag-of-words process with large number of centroids, tend to be sparse holding zeros in many dimensions. Therefore, these dimensions do not have more discriminative power than the rest of dimensions of high value cardinalities [11]. By considering the dimensions value cardinalities in the search strategy, MSIDX clearly outperformed competitive similarity search strategies.…”

Section: Introductionmentioning

confidence: 99%

“…descriptor vectors. Recently, the MSIDX method exploited a new key factor of the image descriptor vectors, namely the dimensions value cardinalities [11]. The dimensions value cardinalities represent the number of discrete values occurred in a specific dimension throughout a database of image descriptor vectors.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Parallel similarity search based on the dimensions value cardinalities of image descriptor vectors

Rafailidis

Manolopoulos

2015

Proceedings of the 30th Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a parallel similarity search strategy based on the dimensions value cardinalities, an inherit characteristic of image descriptor vectors. Our strategy has low preprocessing requirements by dividing the computational cost of the preprocessing steps into several machines and locating the descriptors with similar dimensions value cardinalities logically close. Additionally, an efficient parallel query processing algorithm is proposed, where the dimensions of image descriptors are prioritized in the searching strategy, assuming that dimensions of high value cardinalities have more discriminative power than the dimensions of low ones. In our experiments with publicly available datasets of 80 million and 1 billion images, we show that the proposed method outperforms state-of-the-art parallel similarity search strategies, in terms of preprocessing cost, search time and accuracy. Finally, we made our source code publicly available.

show abstract

Section: Similarity Search Based On Dvcmentioning

confidence: 99%

Section: Similarity Search Based On Dvcmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Parallel similarity search based on the dimensions value cardinalities of image descriptor vectors

Rafailidis

Manolopoulos

2015

Proceedings of the 30th Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…The theory of digital image representation and method used for extraction of color feature gives color feature set of an image. Among the color feature extraction algorithms color histogram (CH) is one of the most frequently used color feature algorithm, this was proposed by Swain and Ballard [1] [8], to use the CH for retrieval. CH has low complexity of computation and it will not change on translation and rotation, as well as robust to scaling and occlusions [2] [9].…”

Section: Introductionmentioning

confidence: 99%

Content-based Image Retrieval using Conflation of Wavelet Transformation and CIECAM02 Color Histogram

Prudviraj¹,

Wadhvani²,

Gyanchandani³

2015

IJCA

View full text Add to dashboard Cite

In this paper, a novel image retrieval technique based on the combination of Haar wavelet transformation and CIECAM02 color histogram (CH) have been proposed. In color based image retrieval, color histogram is one of the most repeatedly used image features and it is used at a great extent in content-based image retrieval (CBIR) systems as a significant color feature. The color histogram unchanged by translation and rotation. The local characteristics and texture features of an image are extracted by wavelet transformation. On conflation of wavelet transformation and color histogram new algorithm has been proposed. One may select a query image perceived to be similar to the visualized target image. A set of images similar to the query is then returned from the database. The final experimental results show that the proposed technique gives better performance than the other schemes, in terms of retrieval time. General TermsImage retrieval, color extraction methods and texture extraction methods. KeywordsColor Histogram, Haar wavelet Transformation, Content-based image retrieval (CBIR), Color histogram, CIECAM02.

show abstract

Skyline-based dissimilarity of images

et al. 2019

View full text Add to dashboard Cite

MSIDX: Multi-Sort Indexing for Efficient Content-Based Image Search and Retrieval

Cited by 27 publications

References 45 publications

Parallel similarity search based on the dimensions value cardinalities of image descriptor vectors

Parallel similarity search based on the dimensions value cardinalities of image descriptor vectors

Content-based Image Retrieval using Conflation of Wavelet Transformation and CIECAM02 Color Histogram

Skyline-based dissimilarity of images

Contact Info

Product

Resources

About