Minor Component Analysis (MCA) Applied to Image Classification in CBIR Systems

Janković, Marko; Zajic, Goran; Radosavljević, Vladan; Kojić, Nenad; Reljin, Nikola; Rudinac, Maja; Rudinac, Stevan; Reljin, Branimir

doi:10.1109/neurel.2006.341164

Cited by 4 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Results obtained by using 25-component FVs are compared with results when using the same CBIR engine but with the full-length FV of 556 components inspired by MPEG-7 [19][20], and with the system using feature vector reduction of about 90% [21,22]. Full-length FVs describe the color (dominant colors in HSV and YCbCr spaces, with 32 components each), color histogram (HSV, 164 components, and YCbCr, 177 components), histogram of line directions (73 components), Gabor texture features (62 coordinates) and gray-level co-occurrence matrix (16 components).…”

Section: Cbir System Evaluation and Resultsmentioning

confidence: 99%

“…We analyzed the retrieval with full-length FVs [20], retrieval with feature vector reduction [22] of about 90% (using about 50 automatically selected components from a full-length FVs), and the retieval with the proposed 25-component FVs. As noted earlier, retrieval quality is evaluated subjectively, by three independent users, and results are averaged.…”

Section: Cbir System Evaluation and Resultsmentioning

confidence: 99%

“…For texture description we use gray-level cooccurrence matrix coefficients (16 coordinates), as we already have used in [20].…”

Section: Cbir With Small Feature Vectormentioning

confidence: 99%

“…In our previous work we have suggested several methods targeted to accelerate search and retrieval in large image databases, without significant degradation of retrieving accuracy: the adaptive clustering of image database (ACID) system [19], a system which uses minor-component analysis (MCA) [20], and a system with feature vector reduction (FVR) [21,22].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Comparison of CBIR Systems with Different Number of Feature Vector Components

Líu

2007

Second International Workshop on Semantic Media Adaptation and Personalization (SMAP 2007)

Self Cite

View full text Add to dashboard Cite

Content-based image retrieval (CBIR) systems with user relevance feedback are considered. The influence of the type and the number of feature vector (FV) components on the retrieval efficiency was investigated. We compared a CBIR system with a very small number of FV components (only 25 components describing color and texture) with a system with a high-dimensional FV inspired by MPEG-7 (556 coordinates describing color, texture and line directions), as well as with a system using feature vector reduction (FVR) of about 90% (with about 50 FV components from the full-length 556-component FVs). The systems are tested over the annotated Corel 1K and Corel 60K datasets. Simulation results show that a decreased number of FV components does not have significant influence on the quality of image retrieval, while the processing time is reduced compared to CBIR with full-length FV and/or FVR.

show abstract

Section: Cbir System Evaluation and Resultsmentioning

confidence: 99%

Section: Cbir System Evaluation and Resultsmentioning

confidence: 99%

“…For texture description we use gray-level cooccurrence matrix coefficients (16 coordinates), as we already have used in [20].…”

Section: Cbir With Small Feature Vectormentioning

confidence: 99%