Adaptive edge detection method for images

Abstract

The novel two-dimensional (2D) wavelet with anisotropic property and application of it has been presented. Wavelet is constructed in the polar coordinate system to obtain anisotropic properties. A novel edge detection method has been developed with the aid of this wavelet. This method detects gradient jump and than follows along this jump. In this way the number of calculation for edge localization is reduced. Moreover, the presented method is able to detect all edges in an image in multi-scale together with its spatial orientation. Proposed wavelet as well as edge extraction method seems to be new way to edge detection for an image.

Edge detection using fast Bidimensional Empirical Mode Decomposition and mathematical morphology

“…In the late 1980s, the Canny edge detection algorithm started to be known as the optimal edge detector [5]. Recently, another novel twodimensional (2D) wavelet with anisotropic property and its application in edge detection have been presented [6].…”

A Comparison of Subpixel Edge Detection and Correlation Algorithms for the Measurement of Sprays

“…Edge precision within the filter width (typically several pixels) is lost. In more modern algorithms such as the adaptive wavelet [15], trigonometric interpolation and adaptive thresholding provide subpixel precision but require low levels of local noise. In the optimised algorithm presented in this paper, subpixel precision is achieved by employing the subpixel curve-fitting and automatic thresholding typical to modern wavelet techniques, but in the form of a Gaussian fit typical of filtering algorithms, permitting reduced noise and defocus sensitivity.…”

“…3. Subpixel location within this intensity profile can be inferenced from local pixel intensity changes and curve-fitting between pixel boundaries [15,22]. To do so, knowledge of the form of the normal intensity profile is necessary.…”

“…These algorithms are mature and offer excellent noise rejection but are precise only to within the filter size; typically several pixels. More modern approaches to edge detection include the adaptive wavelet [15] or phase congruency [16] algorithms. These are more advanced, and employ wavelet fitting and trigonometric interpolation to provide a much higher level of edge precision under varying lighting conditions.…”