Curve Matching, Time Warping, and Light Fields: New Algorithms for Computing Similarity between Curves

Efrat, Alon; Fan, Quanfu; Venkatasubramanian, Suresh

doi:10.1007/s10851-006-0647-0

Cited by 136 publications

(78 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The Fréchet distance (FD) takes the continuity of shapes into account and is a measure of the similarity between two curves. The Fréchet distance or the "dog-man" distance is the minimum length of a leash required to connect a dog and its owner, that walk on two separate curves, without backtracking along their ways [13]. Both of them may vary their speed, and even they can stop at an arbitrary location and for an unpredictable time, but they are not allowed to move backwards.…”

Section: Methods For Comparative Analysismentioning

confidence: 99%

Diagnostic accuracy of ovarian cyst segmentation in B-mode ultrasound images

Bibicu¹,

Moraru²,

Stratulat³

2013

AIP Conference Proceedings

View full text Add to dashboard Cite

Abstract. Cystic and polycystic ovary syndrome is an endocrine disorder affecting women in the fertile age. The Moore Neighbor Contour, Watershed Method, Active Contour Models, and a recent method based on Active Contour Model with Selective Binary and Gaussian Filtering Regularized Level Set (ACM&SBGFRLS) techniques were used in this paper to detect the border of the ovarian cyst from echography images. In order to analyze the efficiency of the segmentation an original computer aided software application developed in MATLAB was proposed. The results of the segmentation were compared and evaluated against the reference contour manually delineated by a sonography specialist. Both the accuracy and time complexity of the segmentation tasks are investigated. The Fréchet distance (FD) as a similarity measure between two curves and the area error rate (AER) parameter as the difference between the segmented areas are used as estimators of the segmentation accuracy. In this study, the most efficient methods for the segmentation of the ovarian were analyzed cyst. The research was carried out on a set of 34 ultrasound images of the ovarian cyst.

show abstract

Section: Methods For Comparative Analysismentioning

confidence: 99%

Diagnostic accuracy of ovarian cyst segmentation in B-mode ultrasound images

Bibicu¹,

Moraru²,

Stratulat³

2013

AIP Conference Proceedings

View full text Add to dashboard Cite

show abstract

“…DTW algorithm [4] has being proved to be extremely efficient for detecting similar shapes with different phases. Given two curves X = (x 1 ; x 2 ; ... …”

Section: Initial Mapping Constructing With Dynamic Time Warpingmentioning

confidence: 99%

“…One thing worth mentioning is that to ensure the mapping between 2 curves is diffeomorphic, we locally turbulent the result if two vertices i and i + 1 were mapped to one vertex j. For more detail about DTW algorithm we refer readers to [4]. …”

Section: Initial Mapping Constructing With Dynamic Time Warpingmentioning

confidence: 99%

Hyperbolic Harmonic Brain Surface Registration with Curvature-Based Landmark Matching

Zeng

Wang

et al. 2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Brain Cortical surface registration is required for inter-subject studies of functional and anatomical data. Harmonic mapping has been applied for brain mapping, due to its existence, uniqueness, regularity and numerical stability. In order to improve the registration accuracy, sculcal landmarks are usually used as constraints for brain registration. Unfortunately, constrained harmonic mappings may not be diffeomorphic and produces invalid registration. This work conquers this problem by changing the Riemannian metric on the target cortical surface to a hyperbolic metric, so that the harmonic mapping is guaranteed to be a diffeomorphism. With landmarks delineated as boundary condition, it is possible to be integrated with landmark matching. The computational algorithms are based on the Ricci flow method and hyperbolic heat diffusion. Experimental results demonstrate that, by changing the Riemannian metric, the registrations have higher qualities in terms of landmark alignment, curvature matching, area distortion and overlapping of region of interests.

show abstract

“…Dynamic Time Warping (DTW) was initially introduced to recognize spoken words [7]. Since then it has been used and proved useful in different applications such as: handwriting recognition [6,8], signature verification [9][10][11][12], fingers print verification [13], face recognition [14,15], speech and gesture recognition [16], protein structure [17], gene expression [18,19], chromosome classification [20], object detection and classification [21,22], curve matching [23,24] control system [25], image and shape matching [24,26] database and data mining [27-29, 31, 32, 34-37), brain activity classification [38], voice control [39] and so on. The DTW algorithm uses a dynamic programming technique to align time series with a given template so that a total distance measure is minimized [25].…”

Section: Visiting Dynamic Time Warping Reviewmentioning

confidence: 99%

Robust Face Localization Using Dynamic Time Warping Algorithm

Majed¹

2011

Reviews, Refinements and New Ideas in Face Recognition

View full text Add to dashboard Cite

Curve Matching, Time Warping, and Light Fields: New Algorithms for Computing Similarity between Curves

Cited by 136 publications

References 28 publications

Diagnostic accuracy of ovarian cyst segmentation in B-mode ultrasound images

Diagnostic accuracy of ovarian cyst segmentation in B-mode ultrasound images

Hyperbolic Harmonic Brain Surface Registration with Curvature-Based Landmark Matching

Robust Face Localization Using Dynamic Time Warping Algorithm

Contact Info

Product

Resources

About