Region Queries without Segmentation for Image Retrieval by Content

Malki, Jamal; Boujemaa, Nozha; Nastar, Chahab; Winter, A.

doi:10.1007/3-540-48762-x_15

Cited by 34 publications

(31 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For example, Minka and Picard [13] divide the image into small sub-images, which are then gathered using a flexible learning algorithm based on the competition within a 'society of models'. Malki et al [20] introduce a multi-resolution signature based on a quadtree, in which every localised and structured region is then indexed by a feature histogram.…”

Section: Image Signatures: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

RETIN: A Content-Based Image Indexing and Retrieval System

Fournier

Cord

Philipp‐Foliguet

2001

Pattern Analysis & Applications

View full text Add to dashboard Cite

This paper presents RETIN, a new system for automatic image indexing and interactive content-based image retrieval. The most original aspect of our work rests on the distance computation and its adjustment by relevance feedback. First, during an offline stage, the indexes are computed from attribute vectors associated with image pixels. The feature spaces are partitioned through an unsupervised classification, and then, thanks to these partitions, statistical distributions are processed for each image. During the online use of the system, the user makes an iconic request, i.e. he brings an example of the type of image he is looking for. The query may be global or partial, since the user can reduce his request to a region of interest. The comparison between the query distribution and that of every image in the collection is carried out by using a weighted dissimilarity function which manages the use of several attributes. The results of the search are then refined by means of relevance feedback, which tunes the weights of the dissimilarity measure via user interaction. Experiments are then performed on large databases and statistical quality assessment shows the good properties of RETIN for digital image retrieval. The evaluation also shows that relevance feedback brings flexibility and robustness to the search.

show abstract

Section: Image Signatures: Related Workmentioning

confidence: 99%

“…Image segmentation is an ill-posed problem, which is why partial request without segmentation, as proposed by Malki et al [20] or Minka and Picard [13], is an effective alternative. The user now specifies the blocks of interest in the stamped query image.…”

Section: Search-by-similarity: Related Workmentioning

confidence: 99%

RETIN: A Content-Based Image Indexing and Retrieval System

Fournier

Cord

Philipp‐Foliguet

2001

Pattern Analysis & Applications

View full text Add to dashboard Cite

show abstract

“…In this case, each image is represented by a full balanced tree. Each tree node (internal or leaf) stores the features of the corresponding image segment, for example a color histogram (Gupta et al, 1997;Lu et al, 1994;Lin et al, 2001), (Malki et al, 1999) or several relevant JPEG coefficients (Climer et al, 2002). Figure 6 gives an example of such a structure: each quadtree node contains the color histogram of the corresponding image segment.…”

Section: Multi-level Feature Vectorsmentioning

confidence: 97%

“…To avoid this kind of limitation, the query pattern can be transformed into several query patterns, each one representing a geometric transformation of the initial one. For example, in Malki et al (1999), the user selects several segments in a query image represented by a quadtree (see (a) to (c) in Fig. 7).…”

Section: P -Distancementioning

confidence: 99%

“…Several approaches (Climer et al, 2002;Gupta et al, 1997;Jomier et al, 2005;Kim et al, 2000;Lin et al, 2001;Lu et al, 1994;Luo et al, 2003;Malki et al, 1999;Wang Z et al, 2003) represent each image by a tree having a fixed number of levels (usually a full 3-level tree) to content-based image retrieval. In this case, each image is represented by a full balanced tree.…”

Section: Multi-level Feature Vectorsmentioning

confidence: 99%

See 1 more Smart Citation

Δ-distance: A family of dissimilarity metrics between images represented by multi-level feature vectors

2006

View full text Add to dashboard Cite

This article presents the -distance, a family of distances between images recursively decomposed into segments and represented by multi-level feature vectors. Such a structure is a quad, a quin or a nona-tree resulting from a fixed and arbitrary image partition or from an image segmentation process. It handles positional information of image features (e.g. color, texture or shape). -distance is the generalized form of dissimilarity measures between multi-level feature vectors. Using different weights on tree nodes and different distances between nodes, distances between trees or visual similarity between images can be computed based on the general definition of . In this article, we present three -based distance families: two families of distances between tree structures, called T -distance (T for Tree) and S-distance (S for Segment), and a family of visual distances between images, called V-distance (V for Visual). The V-distance visually compares two images using their tree representation and the other two distances compare the tree structures resulting from image segmentation. Moreover, we show how existing distances between multi-level feature vectors appear to be particular cases of the -distance.Keywords Image database . Distance between quad/quin or nona-trees . Similarity of images . Similarity of image segments . Content-based image retrieval

show abstract