Reconstruction of images from Gabor graphs with applications in facial image processing

Günther, Manuel; Böhringer, Stefan; Wieczorek, Dagmar; Würtz, Rolf P.

doi:10.1142/s0219691315500198

Cited by 6 publications

(3 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The weighting is linear in both the size of the individual heritability and the inverse distance of the center of the feature with the current point. Details of this procedure are given elsewhere 25 , 26 .…”

Section: Methodsmentioning

confidence: 99%

Ensemble landmarking of 3D facial surface scans

Jong

Hysi

Spector

et al. 2018

Sci Rep

Self Cite

View full text Add to dashboard Cite

Landmarking of 3D facial surface scans is an important analysis step in medical and biological applications, such as genome-wide association studies (GWAS). Manual landmarking is often employed with considerable cost and rater dependent variability. Landmarking automatically with minimal training is therefore desirable. We apply statistical ensemble methods to improve automated landmarking of 3D facial surface scans. Base landmarking algorithms using features derived from 3D surface scans are combined using either bagging or stacking. A focus is on low training complexity of maximal 40 training samples with template based landmarking algorithms that have proved successful in such applications. Additionally, we use correlations between landmark coordinates by introducing a search strategy guided by principal components (PCs) of training landmarks. We found that bagging has no useful impact, while stacking strongly improves accuracy to an average error of 1.7 mm across all 21 landmarks in this study, a 22% improvement as compared to a previous, comparable algorithm. Heritability estimates in twin pairs also show improvements when using facial distances from landmarks. Ensemble methods allow improvement of automatic, accurate landmarking of 3D facial images with minimal training which is advantageous in large cohort studies for GWAS and when landmarking needs change or data quality varies.

show abstract

Section: Methodsmentioning

confidence: 99%

Ensemble landmarking of 3D facial surface scans

Jong

Hysi

Spector

et al. 2018

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…We used methods described previously to analyze the data ( Boehringer et al, 2011a ; Balliu et al, 2014 ; Günther et al, 2015 ). In short, landmarks were manually placed on frontal photographs of patients and controls.…”

Section: Methodsmentioning

confidence: 99%

Distinctive facial features in idiopathic Moyamoya disease in Caucasians: a first systematic analysis

Kraemer

Huynh

Wieczorek

et al. 2018

PeerJ

Self Cite

View full text Add to dashboard Cite

BackgroundCraniofacial dysmorphic features are morphological changes of the face and skull which are associated with syndromic conditions. Moyamoya angiopathy is a rare cerebral vasculopathy that can be divided into Moyamoya syndrome, which is associated or secondary to other diseases, and into idiopathic Moyamoya disease. Facial dysmorphism has been described in rare genetic syndromes with associated Moyamoya syndrome. However, a direct relationship between idiopathic Moyamoya disease with dysmorphic facial changes is not known yet.MethodsLandmarks were manually placed on frontal photographs of the face of 45 patients with bilateral Moyamoya disease and 50 matched controls. After procrustes alignment of landmarks a multivariate, penalized logistic regression (elastic-net) was performed on geometric features derived from landmark data to classify patients against controls. Classifiers were visualized in importance plots that colorcode importance of geometric locations for the classification decision.ResultsThe classification accuracy for discriminating the total patient group from controls was 82.3% (P-value = 6.3×10−11, binomial test, a-priori chance 50.2%) for an elastic-net classifier. Importance plots show that differences around the eyes and forehead were responsible for the discrimination. Subgroup analysis corrected for body mass index confirmed a similar result.DiscussionResults suggest that there is a resemblance in faces of Caucasian patients with idiopathic Moyamoya disease and that there is a difference to matched controls. Replication of findings is necessary as it is difficult to control all residual confounding in study designs such as ours. If our results would be replicated in a larger cohort, this would be helpful for pathophysiological interpretation and early detection of the disease.

show abstract

“…In Ref. 15, a method of image reconstruction is presented by an interpolation rule and an iterative algorithm from Gabor graphs. A recursive algorithm based on random noise injection in the unobserved space of the image and denoising using a spatial adaptive filter is proposed in Ref.…”

Section: Literature Review On the Existing Work And Limitationsmentioning

confidence: 99%

On curvelet CS reconstructed MR images and GA-based fuzzy conditional entropy maximization for segmentation

Roy¹,

Maity

2019

Int. J. Wavelets Multiresolut Inf. Process.

View full text Add to dashboard Cite

In many practical situations, magnetic resonance imaging (MRI) needs reconstruction of images at low measurements, far below the Nyquist rate, as sensing process may be very costly and slow enough so that one can measure the coefficients only a few times. Segmentation of such subsampled reconstructed MR images for medical analysis and diagnosis becomes a challenging task due to the inherent complex characteristics of the MR images. This paper considers reconstruction of MR images at compressive sampling (or compressed sensing (CS)) paradigm followed by its segmentation in an integrated platform. Image reconstruction is done from incomplete measurement space with random noise injection iteratively. A weighted linear prediction is done for the unobserved space followed by spatial domain denoising through adaptive recursive filtering. The reconstructed images, however, suffer from imprecise and/or missing edges, boundaries, lines, curvatures etc. and residual noise. Curvelet transform (CT) is purposely used for removal of noise and for edge enhancement through hard thresholding and suppression of approximate subbands, respectively. Then a fuzzy entropy-based clustering, using genetic algorithms (GAs), is done for segmentation of sharpen MR Image. Extensive simulation results are shown to highlight performance improvement of both image reconstruction and segmentation of the reconstructed images along with relative gain over the existing works.

show abstract

Reconstruction of images from Gabor graphs with applications in facial image processing

Cited by 6 publications

References 22 publications

Ensemble landmarking of 3D facial surface scans

Ensemble landmarking of 3D facial surface scans

Distinctive facial features in idiopathic Moyamoya disease in Caucasians: a first systematic analysis

On curvelet CS reconstructed MR images and GA-based fuzzy conditional entropy maximization for segmentation

Contact Info

Product

Resources

About