Study on Algorithms of Graphic Element Recognition for Precise Vectorization of Industrial Computed Tomographic Image

Liu, Fenglin

doi:10.2174/1874155x01105010043

Cited by 2 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We then compute T , using the interquartile range (IQR): T = Q 3+ W ×IQR where W =1.5 as is common in outlier detection (McGill et al, 1978) and IQR= Q 3− Q 1, Q 1 and Q 3 are, respectively, the first and the third quartiles of the distribution of |A| . Thus, the condition max( |A| ) <T will provide a stop criterion for the main loop.…”

Section: Methodsmentioning

confidence: 99%

Automatic Biological Cell Counting Using a Modified Gradient Hough Transform

et al. 2017

View full text Add to dashboard Cite

We present a computational method for pseudo-circular object detection and quantitative characterization in digital images, using the gradient accumulation matrix as a basic tool. This Gradient Accumulation Transform (GAT) was first introduced in 1992 by Kierkegaard and recently used by Kaytanli & Valentine. In the present article, we modify the approach by using the phase coding studied by Cicconet, and by adding a "local contributor list" (LCL) as well as a "used contributor matrix" (UCM), which allow for accurate peak detection and exploitation. These changes help make the GAT algorithm a robust and precise method to automatically detect pseudo-circular objects in a microscopic image. We then present an application of the method to cell counting in microbiological images.

show abstract

Section: Methodsmentioning

confidence: 99%

Automatic Biological Cell Counting Using a Modified Gradient Hough Transform

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Vectorized contours are composed of several basic primitives, such as circles, arcs and segments. In addition, some algorithms have been presented to identify primitives from contours of slices [ 19 , 20 ].…”

Section: Introductionmentioning

confidence: 99%

A shape-based inter-layer contours correspondence method for ICT-based reverse engineering

et al. 2017

View full text Add to dashboard Cite

The correspondence of a stack of planar contours in ICT (industrial computed tomography)-based reverse engineering, a key step in surface reconstruction, is difficult when the contours or topology of the object are complex. Given the regularity of industrial parts and similarity of the inter-layer contours, a specialized shape-based inter-layer contours correspondence method for ICT-based reverse engineering was presented to solve the above problem based on the vectorized contours. In this paper, the vectorized contours extracted from the slices consist of three graphical primitives: circles, arcs and segments. First, the correspondence of the inter-layer primitives is conducted based on the characteristics of the primitives. Second, based on the corresponded primitives, the inter-layer contours correspond with each other using the proximity rules and exhaustive search. The proposed method can make full use of the shape information to handle industrial parts with complex structures. The feasibility and superiority of this method have been demonstrated via the related experiments. This method can play an instructive role in practice and provide a reference for the related research.

show abstract

Study on Algorithms of Graphic Element Recognition for Precise Vectorization of Industrial Computed Tomographic Image

Cited by 2 publications

References 4 publications

Automatic Biological Cell Counting Using a Modified Gradient Hough Transform

Automatic Biological Cell Counting Using a Modified Gradient Hough Transform

A shape-based inter-layer contours correspondence method for ICT-based reverse engineering

Contact Info

Product

Resources

About