Robust image salient regional extraction and matching based on DoGSS-MSERs

Zhang, Mengmeng; Li, Zeming; Bai, Huihui; Sun, Yan

doi:10.1016/j.ijleo.2013.09.007

Cited by 6 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They have concentrated on the algorithmic details of MSER and concurrent construction of the union finding forest and the components' (extremal region) tree. There have also been lots of extensions over the original version of MSER, as those in [40]- [46]. In [43], an integrated algorithm, ED-MSER, is proposed which combines MSER [18] with SIFT [47] and a filtering strategy.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Extremal Regions Detection Guided by Maxima of Gradient Magnitude

Faraji

Shanbehzadeh

Nasrollahi

et al. 2015

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

A problem of computer vision applications is to detect regions of interest under different imaging conditions. The state-of-the-art maximally stable extremal regions (MSERs) detects affine covariant regions by applying all possible thresholds on the input image, and through three main steps including: (1) making a component tree of extremal regions' evolution; (2) obtaining region stability criterion; and (3) cleaning up. The MSER performs very well, but, it does not consider any information about the boundaries of the regions, which are important for detecting repeatable extremal regions. We have shown in this paper that employing prior information about boundaries of regions results in a novel region detector algorithm that not only outperforms MSER, but avoids the MSER's rather complicated steps of enumeration and the cleaning up. To employ the information about the region boundaries, we introduce maxima of gradient magnitudes (MGMs) which are shown to be points that are mostly around the boundaries of the regions. Having found the MGMs, the method obtains a global criterion for each level of the input image which is used to find extremum levels (ELs). The found ELs are then used to detect extremal regions. The proposed algorithm which is called extremal regions of extremum levels (EREL) has been tested on the public benchmark data set of Mikolajczyk. The obtained experimental results show that the inclusion of region boundaries through MGMs, results in a detector that detects regions with high repeatability scores and is more robust against noise compared with MSER.

show abstract

Section: Introductionmentioning

confidence: 99%

“…Enclosed regions consist of External Enclosing Regions (EER) and Internal Enclosed Regions (IER). Recently, MSER has been adapted to work with scale-space theory as for example in [45], [46], [48]. Forseen and Lowe [48] introduced a multi-resolution version of MSER and used it to construct a descriptor for detected extremal regions.…”

Section: Introductionmentioning

confidence: 99%

Extremal Regions Detection Guided by Maxima of Gradient Magnitude

Faraji

Shanbehzadeh

Nasrollahi

et al. 2015

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

show abstract

“…They have concentrated on the algorithmic details of MSER and concurrent construction of the union finding forest and the components' (extremal regions) tree. On the other hand, there are lots of extensions over the original version of MSER, as those in [6,7,8,9,10,11,12]. In [9], an integrated algorithm, ED-MSER, is proposed which combines MSER [2] with SIFT [13] and a filtering strategy.…”

Section: Introductionmentioning

confidence: 99%

“…Enclosed regions consist of External Enclosing Regions (EER) and Internal Enclosed Regions (IER). Recently, MSER has been adapted to work with scale-space theory in [14,11,12]. Forseen and Lowe [14] introduced a multiresolution version of MSER and used it to construct a descriptor for detected extremal regions.…”

Section: Introductionmentioning

confidence: 99%

“…Forseen and Lowe [14] introduced a multiresolution version of MSER and used it to construct a descriptor for detected extremal regions. In [12], MSER is employed on the Difference of Gaussian (DoG). The stable regions are selected if the barycenter of a region is surrounded by at least ten barycenters of other regions in its adjacent scales.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation