Vibration measurement of a rotating cylindrical structure using subpixel-based edge detection and edge tracking

Edge detection is a fundamental task in many computer vision applications. In this paper, we propose a novel multiscale edge detection approach based on the nonsubsampled contourlet transform (NSCT): a fully shift-invariant, multiscale, and multidirection transform. Indeed, unlike traditional wavelets, contourlets have the ability to fully capture directional and other geometrical features for images with edges. Firstly, compute the NSCT of the input image. Secondly, theK-means clustering algorithm is applied to each level of the NSCT for distinguishing noises from edges. Thirdly, we select the edge point candidates of the input image by identifying the NSCT modulus maximum at each scale. Finally, the edge tracking algorithm from coarser to finer is proposed to improve robustness against spurious responses and accuracy in the location of the edges. Experimental results show that the proposed method achieves better edge detection performance compared with the typical methods. Furthermore, the proposed method also works well for noisy images.

Download Full-text

Multiscale edge detection based on Gaussian smoothing and edge tracking

Knowledge-Based Systems ◽

10.1016/j.knosys.2013.01.026 ◽

2013 ◽

Vol 44 ◽

pp. 101-111 ◽

Cited By ~ 57

Author(s):

C. Lopez-Molina ◽

B. De Baets ◽

H. Bustince ◽

J. Sanz ◽

E. Barrenechea

Keyword(s):

Edge Detection ◽

Gaussian Smoothing ◽

Edge Tracking

Download Full-text

Difference mode EXELFS with parallel-detection EELS

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100130948 ◽

1992 ◽

Vol 50 (2) ◽

pp. 1262-1263

Author(s):

Michael K. Kundmann ◽

Ondrej L. Krivanek

Keyword(s):

Edge Detection ◽

Important Distinction ◽

Edge Structure ◽

Parallel Detection ◽

Gain Variation ◽

Ideal Solutions ◽

Detector Cell ◽

The Difference ◽

Elemental Detection ◽

Partial Correction

Parallel detection has greatly improved the elemental detection sensitivities attainable with EELS. An important element of this advance has been the development of differencing techniques which circumvent limitations imposed by the channel-to-channel gain variation of parallel detectors. The gain variation problem is particularly severe for detection of the subtle post-threshold structure comprising the EXELFS signal. Although correction techniques such as gain averaging or normalization can yield useful EXELFS signals, these are not ideal solutions. The former is a partial throwback to serial detection and the latter can only achieve partial correction because of detector cell inhomogeneities. We consider here the feasibility of using the difference method to efficiently and accurately measure the EXELFS signal.An important distinction between the edge-detection and EXELFS cases lies in the energy-space periodicities which comprise the two signals. Edge detection involves the near-edge structure and its well-defined, shortperiod (5-10 eV) oscillations. On the other hand, EXELFS has continuously changing long-period oscillations (∼10-100 eV).

Download Full-text