## Detection of DNA Filaments in Fluorescent Microscopy Using Feature-adapted Beamlet Transform

Citations: | 1 - 0 self |

### BibTeX

@MISC{Berlemont_detectionof,

author = {Sylvain Berlemont and Aaron Bensimon and Jean-christophe Olivo-marin},

title = {Detection of DNA Filaments in Fluorescent Microscopy Using Feature-adapted Beamlet Transform},

year = {}

}

### OpenURL

### Abstract

This paper presents a new method for computing the Feature-adapted Beamlet transforms [1] in a fast and accurate way. This transform can be used for detecting features running along lines or piecewise constant curves. The main contribution of this paper is to unify the Fast Slant Stack method, introduced in [2], with linear filtering technique in order to implement the Feature-adapted Beamlet transform. If the desired feature detector is chosen to belong to the class of steerable filters, our method can be achieved in O(N log(N)), where N = n 2 is the number of pixels. This new method leads to an efficient implementation of Feature-adapted Beamlet transforms, that outperforms our previous works [1] both in terms of accuracy and speed. Our method has been developed in the context of biological imaging to detect DNA filaments in fluorescent microscopy.

### Citations

3098 | A computational approach to edge detection
- Canny
- 1986
(Show Context)
Citation Context ...o detect curvilinear objects is to track locally the feature of the curve-profile; linear filtering or template matched filtering are well-known techniques for doing so. Classical Canny edge detector =-=[6]-=- is based on such linear filtering techniques. It involves the computation of inner-products with shifted and/or rotated versions of the feature template at every point in the image. High response at ... |

847 | The design and use of steerable filters
- Freeman, Adelson
- 1991
(Show Context)
Citation Context ... Fourier transform as many times as the number of θ’s, i.e. 2n times. To overcome this limitation, we propose to study the special case where h is selected to be within the class of steerable filters =-=[9]-=-. We can write h θ as a linear combination of basis filters: h θ (x, y) = M� φj(θ)h θj (x, y), (7) j=1 where φj’s are interpolation functions that only depend on θ and the basis filters h θj ’s are in... |

281 | D.L.Donoho Curvelets: A Surprisingly Effective Nonadaptive Reprentation for Objects with Edges
- Candes
(Show Context)
Citation Context ...ote sensing images, human vasculature in medical imaging or DNA filaments in biological microscopy. Edges can be also percieved as curvilinear objects since they describe discontinuities along curves =-=[4]-=-. Therefore, detection of such features is probably one of the most important step toward automatic object recognition and shape analysis. Commonly, curvilinear objects are considered as 1-dimensional... |

55 | Beamlets and multiscale image analysis
- Donoho, Huo
- 2002
(Show Context)
Citation Context ...aluation of line integrals upon the whole set of segments is practically infeasible due to the computational burden. One of the methodologies proposed to address this problem is the Beamlet transform =-=[7]-=-. It defines a set of dyadically organized line segments (beamlets) occupying a range of dyadic locations and scales, and spanning a full range of orientations. The collection of beamlets has a O(N lo... |

49 | Fast slant stack: A notion of radon transform for data in a cartesian grid which is rapidly computible, algebraically exact, geometrically faithful and invertible
- Averbuch, Coifman, et al.
- 2001
(Show Context)
Citation Context ...ate way. This transform can be used for detecting features running along lines or piecewise constant curves. The main contribution of this paper is to unify the Fast Slant Stack method, introduced in =-=[2]-=-, with linear filtering technique in order to implement the Feature-adapted Beamlet transform. If the desired feature detector is chosen to belong to the class of steerable filters, our method can be ... |

47 | Design of steerable filters for feature detection using canny-like criteria
- Jacob, Unser
- 2004
(Show Context)
Citation Context ...the line-profile has been selected within the class of steerable filters. In the next paragraph, we discuss about the choice of such filters. θ 5sB. Choice of filter h Recently, detectors designed in =-=[11]-=- out-performed classical feature detectors like Canny edge detector or 2 nd derivative of gaussian-based filter for ridge detection. Authors proposed a general approach for the design of 2D feature de... |

20 |
Gaussian approximations of fluorescence microscope point-spread function models
- Zhang, Zerubia, et al.
- 2007
(Show Context)
Citation Context ...ty of the filament (we suppose our filaments being relativaly free of intensity oscillations) and ρ is the PSF of the microscope. Approximation of PSF for different type of microscope can be found in =-=[5]-=-. One way to detect curvilinear objects is to track locally the feature of the curve-profile; linear filtering or template matched filtering are well-known techniques for doing so. Classical Canny edg... |

14 |
Fast calculation of multiple line integrals
- Brandt, Dym
- 1999
(Show Context)
Citation Context ...d by a discrete lattice I = I(u, v) : −n/2 ≤ u, v < n/2. There are two distinct approaches to compute equation (2) efficiently on digital arrays. The first one is the multiscale approach presented in =-=[8]-=-. This approach splits the image domain by recursive partitioning into dyadic squares. Radon coefficients are computed by recombining those which have been already computed on smaller squares. The maj... |

9 | Alignment and sensitive detection of DNA by a moving interface - Bensimon, Simon, et al. - 1994 |

2 |
Detection of curvilinear objects in biological noisy image using feature-adapted fast slant stack
- Berlemont, Bensimon, et al.
- 2007
(Show Context)
Citation Context ..., the 2n numbers R[I ∗ h θ ](t, θ), −n ≤ t < n, are a 1-dimensional discrete Fourier transform pair with the 2n numbers M� j=1 φj(θ) � I ∗ hθj(π k n k tan(θ), π ), −n ≤ k < n. n The proof is given in =-=[10]-=-. We call this transform the Feature-adapted Fast Slant Stack. Thanks to this result, we compute equation (6) for every angle, as follows: first we convolve the image as many times as the number of ba... |

1 |
Detection of curvilinear objects in noisy image using feature-adapted beamlet transform
- Berlemont, Bensimon, et al.
- 2007
(Show Context)
Citation Context ...is, France 3 Genomic Vision, Paris Biotech Santé, 24, rue du faubourg Saint Jacques, 75014 Paris, France Abstract This paper presents a new method for computing the Feature-adapted Beamlet transforms =-=[1]-=- in a fast and accurate way. This transform can be used for detecting features running along lines or piecewise constant curves. The main contribution of this paper is to unify the Fast Slant Stack me... |

1 |
Wedgelets: near minmax estimation of edges,” The
- Donoho
- 1999
(Show Context)
Citation Context ...es a list of beamlets that best represent curvilinear objects carrying a specific line-profile in an image. This method is based on a multiscale coefficient thresholding technique directly taken from =-=[12]-=- so we refer the reader to this paper for more details. A. Algorithm A Recursive Dyadic Partition (RDP) of the image domain is any partition, starting from the whole image domain, obtained by recursiv... |