Zero-crossings in a second-derivative-of-Gaussian filtered image is a well-known edge location criterion. Examples are the Laplacian and directional second derivatives such as the second derivative in the gradient dir...
详细信息
Zero-crossings in a second-derivative-of-Gaussian filtered image is a well-known edge location criterion. Examples are the Laplacian and directional second derivatives such as the second derivative in the gradient direction (SDGD). Derivative operators can easily be implemented as a convolution with a derivative of a Gaussian. Gaussian filtering displaces the half height isophote towards smaller edge radii (inwards for convex edges and outwards for concave edges). The Difference-of-Gaussians (DoG) filters are similar to the Laplacian-of-Gaussian and exert an edge shift to larger edge radii (outwards). This paper introduces Multiple-of-Gaussians filters with reduced curvature-based error. Using N Gaussians (N>2) we reduce edge shifts to a fraction (1/(2N-3)) of the ones produced by a similar Laplacian-of-Gaussian filter.
The probability density functions of Freeman codes are given under the assumption that the curve segment may be approximated by a line segment in a grid column (or grid row). The a priori probability of odd or even Fr...
The probability density functions of Freeman codes are given under the assumption that the curve segment may be approximated by a line segment in a grid column (or grid row). The a priori probability of odd or even Freeman codes is calculated with these p.d.f.'s. The variance present in the curvature measurement based on Freeman codes is computed. Moreover the p.d.f.'s of curve segment lengths associated with odd and even codes are determined.
A novel method for vectorizing line drawing images is presented. The method is based on a sequence of a standard vectorization algorithm and maximum threshold morphology, which can be iterated until a fitting criterio...
详细信息
A novel method for vectorizing line drawing images is presented. The method is based on a sequence of a standard vectorization algorithm and maximum threshold morphology, which can be iterated until a fitting criterion is met. First, ''anchor points'' are found in a coarse vectorization. The position of each anchor point is corrected by morphological operations, after which it is fixed. Next, the vectorization is refined between the anchor points. The method is adaptive because the original input image is used for correcting and improving the vectorization. Postprocessing modules can be added to anticipate specific properties of the line drawings to be vectorized. The vectorization method is evaluated using line drawings differing in scale and resolution. (C) 1997 Academic Press.
For a class of image measures a scheme is given to obtain new Nyquist rates that guarantee exact measuring results without analog reconstruction. Gradient-power weighting is a compatible alternative for binary contour...
详细信息
For a class of image measures a scheme is given to obtain new Nyquist rates that guarantee exact measuring results without analog reconstruction. Gradient-power weighting is a compatible alternative for binary contouring. Generalized bending energy is introduced as a practical example.
Enhancement of banding patterns in integrated density profiles is widely accepted as a necessary step in processing human chromosome images. In this paper a filter is presented which only enhances the frequencies of i...
Enhancement of banding patterns in integrated density profiles is widely accepted as a necessary step in processing human chromosome images. In this paper a filter is presented which only enhances the frequencies of interest in the banding pattern. A method is given to compute these frequencies with a correction for the contraction state of the metaphase. The presented filter is transparent to the user by its explicitly defined frequency passband and shows a better suppression in the stopbands than filters with prescribed simple weighting functions.
The paper presents an O(N log N)-implementation of t-SNE — an embedding technique that is commonly used for the visualization of high-dimensional data in scatter plots and that normally runs in O(N2). The new impleme...
详细信息
In this paper we propose a recursive implementation of the Gaussian filter. This implementation yields an infinite impulse response filter that has six MADDs per dimension independent of the value of a in the Gaussian...
详细信息
In this paper we propose a recursive implementation of the Gaussian filter. This implementation yields an infinite impulse response filter that has six MADDs per dimension independent of the value of a in the Gaussian kernel. In contrast to the Deriche implementation (1987), the coefficients of our recursive filter have a simple, closed-form solution for a desired value of the Gaussian a. Our implementation is, in general, faster than (1) an implementation based upon direct convolution with samples of a Gaussian, (2) repeated convolutions with a kernel such as the uniform filter, and (3) an FFT implementation of a Gaussian filter.
Applications in automated microscopy and three-dimensional microscopy require careful calibration of the microscope system. This paper presents methods for calibration of the motorized z-axis (focus or optical axis) o...
详细信息
Applications in automated microscopy and three-dimensional microscopy require careful calibration of the microscope system. This paper presents methods for calibration of the motorized z-axis (focus or optical axis) of an automated microscope. Apart from the automated microscope the procedures require a CCD camera and a test slide containing a simple bar pattern. The calibration embraces the following characteristics of the z-axis: (a) measuring the motor step size in nanometres;(b) measuring the mechanical backlash in the focus mechanism of the microscope and (c) measuring the reproducibility and the stability of the focus position over time. The measurements employ focus functions to determine the z-position of the microscope stage.
A novel object operation is proposed, where each point in a destination image contains the value of the 'view angle' with which it can see the object in the source image. Thesholding a view angle image at 180-...
详细信息
A novel object operation is proposed, where each point in a destination image contains the value of the 'view angle' with which it can see the object in the source image. Thesholding a view angle image at 180-degrees gives the convex hull of the objects in the source image.
A unified treatment is presented for determining the distance between two points in a two- or three-dimensional digitized space. Instead of a global Euclidean distance, a distance transformation based upon a sequence ...
详细信息
A unified treatment is presented for determining the distance between two points in a two- or three-dimensional digitized space. Instead of a global Euclidean distance, a distance transformation based upon a sequence of optimal local distances is used. The optimal distance is derived in the context of minimizing the maximum error and the unbiased mean square error. Integer approximations for the local distances are developed for neighborhood sizes of three and five. Minimization is performed over circles and spheres to preserve the symmetries of the neighborhoods. In two dimensions the differences with previously published results are small. The results in three dimensions are new.
暂无评论