Robust Face Recognition through Local Graph Matching

Fazl-Ersi, Ehsan; Zelek, John; Tsotsos, John K.

doi:10.4304/jmm.2.5.31-37

Cited by 9 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we present a heuristic graph matching method, wherein the matching problem is solved in a progressive manner (i.e. cell by cell), by obtaining correspondences from local graphs generated at different time instants (Gold and Rangarajan, 1996; Fazl‐Ersi et al. , 2007).…”

Section: Introductionmentioning

confidence: 99%

Automated tracking of stem cell lineages of Arabidopsis shoot apex using local graph matching

et al. 2009

View full text Add to dashboard Cite

SUMMARYShoot apical meristems (SAMs) of higher plants harbor stem-cell niches. The cells of the stem-cell niche are organized into spatial domains of distinct function and cell behaviors. A coordinated interplay between cell growth dynamics and changes in gene expression is critical to ensure stem-cell homeostasis and organ differentiation. Exploring the causal relationships between cell growth patterns and gene expression dynamics requires quantitative methods to analyze cell behaviors from time-lapse imagery. Although technical breakthroughs in live-imaging methods have revealed spatio-temporal dynamics of SAM-cell growth patterns, robust computational methods for cell segmentation and automated tracking of cells have not been developed. Here we present a local graph matching-based method for automated-tracking of cells and cell divisions of SAMs of Arabidopsis thaliana. The cells of the SAM are tightly clustered in space which poses a unique challenge in computing spatio-temporal correspondences of cells. The local graph-matching principle efficiently exploits the geometric structure and topology of the relative positions of cells in obtaining spatiotemporal correspondences. The tracker integrates information across multiple slices in which a cell may be properly imaged, thus providing robustness to cell tracking in noisy live-imaging datasets. By relying on the local geometry and topology, the method is able to track cells in areas of high curvature such as regions of primordial outgrowth. The cell tracker not only computes the correspondences of cells across spatio-temporal scale, but it also detects cell division events, and identifies daughter cells upon divisions, thus allowing automated estimation of cell lineages from images captured over a period of 72 h. The method presented here should enable quantitative analysis of cell growth patterns and thus facilitating the development of in silico models for SAM growth.

show abstract

Section: Introductionmentioning

confidence: 99%

Automated tracking of stem cell lineages of Arabidopsis shoot apex using local graph matching

et al. 2009

View full text Add to dashboard Cite

show abstract

“…We use the accuracy standard introduced in literature [15] to evaluate the segmentation accuracy of our proposed method. In [15], F is defined as the segmentation accuracy ratio, and P is the Precision ratio in Equation (6), while R represents the Recall ratio in Equation (7). GT N denotes the image edge pixels by artificial segmentation and Det N represents the image edge pixels by the algorithm for image segmentation.…”

Section: Resultsmentioning

confidence: 99%

“…Given sets of segmented cells at different time instants, cell tracking is essentially a kind of vertex matching problem, which has been widely studied. One of the most popular solutions is the local graph matching method proposed in [7,8]. The watershed method and local graph matching method are able to segment and track most of the cells when the cell images are not highly noised, otherwise, there will be an over-segmentation problem.…”

Section: Introductionmentioning

confidence: 99%

Automated Segmentation and Tracking of SAM Cells

Liu

Peng

2014

Communications in Computer and Information Science

View full text Add to dashboard Cite

Abstract. In this paper, we propose an automated segmentation and tracking system for the shoot apical meristem (SAM) cells. Cells are segmented using a mixed filter based watershed segmentation method, which is proved to be very robust and efficient. After segmentation, a Triangle Neighborhood Structure matching method is proposed to track the segmented cells across different time instances. Our tracking method reduces the dependence on neighbors, because we only need two neighbors for any cells for matching while the other local graph matching methods require a much larger number of neighbors. Using our proposed segmentation and tracking system, we are able to track 97% of the plant SAM cells.

show abstract

“…Recognition Rates (EBGM) -Wisskott et al [4] 95.5% (LDA + PCA) -Etemad et al [2] 96.2% (BIC) -Moghaddam et al [3] 94.8% (Boosted Haar) -Jones et al [11] 94.0% (LBP) -Timo et al [12] 97% (LGBPHS) -Zhang et al [13] 98% Fazl-Ersi et al [8] 98% Our proposed method 99.1% Table 1: Comparison of our result on the FERET dataset with the results of several state-of-the-art face recognition methods. When comparing the results, note that almost similar face normalization algorithms were used in all techniques, except [4] and [8].…”

Section: Methodsmentioning

confidence: 99%

“…When comparing the results, note that almost similar face normalization algorithms were used in all techniques, except [4] and [8].…”

Section: Methodsmentioning

confidence: 99%

Local feature analysis for robust face recognition

Fazl-Ersi

Tsotsos

2009

2009 IEEE Symposium on Computational Intelligence for Security and Defense Applications

Self Cite

View full text Add to dashboard Cite

In this paper a novel technique for face recognition is proposed. Using the statistical Local Feature Analysis (LFA) technique, a set of feature points is extracted from each face image, at locations with highest deviations from the statistical expected face. Each feature point is described by a set of Gabor wavelet responses at different frequencies and orientations. A triangle-inequality-based pruning algorithm is developed for fast matching, which automatically chooses a set of key features from the database of model features and uses the pre-computed distances of the keys to the database, along with the triangle inequality, in order to speedily compute lower bounds on the distances from a query feature to the database, and eliminate the unnecessary direct comparisons. Our proposed technique achieves perfect results on the ORL face set and an accuracy rate of 99.1% on the FERET face set, which shows the superiority of the proposed technique over all considered state-of-the-art face recognition methods.

show abstract

Robust Face Recognition through Local Graph Matching

Cited by 9 publications

References 17 publications

Automated tracking of stem cell lineages of Arabidopsis shoot apex using local graph matching

Automated tracking of stem cell lineages of Arabidopsis shoot apex using local graph matching

Automated Segmentation and Tracking of SAM Cells

Local feature analysis for robust face recognition

Contact Info

Product

Resources

About