A Hough transform system based on neural networks

Dempsey, G.L.; McVey, Eugene S.

doi:10.1109/41.170971

Cited by 16 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…18 To improve the accuracy of laser peak detection, the neural network can be used. 19–21 Also, for shiny surfaces, a method for increasing the accuracy and robustness of estimation has been described in Song et al., 22 and the color of the laser can be recognized in an image based on fuzzy systems. 23…”

Section: Calibration Of the Dual-parallel Laser Beam Profilermentioning

confidence: 99%

A new flexible laser beam profiler for the inspection of train wheels

Torabi

Younesian

2020

Proceedings of the Institution of Mechanical Engineers, Part F:

View full text Add to dashboard Cite

In this paper, a flexible laser beam profiler is proposed to easily measure the profile of a train wheel for railway inspection. It only requires two laser beams (together and in parallel) to obtain two three-dimensional point-clouds based on the laser triangulation principle. Either the laser beam profiler or the wheel can be freely moved. The motion need not be known. The wheel profile will be obtained in two steps. First, the wheel axis position and orientation are obtained by minimizing the distance between one of the point-clouds and the other translated point-cloud, and translation is defined as a rotation of any point on the point-cloud around the wheel axis until it lies on the other point-cloud's laser plane. In the second step, the wheel profile is extracted by selecting one of the point-clouds and rotating it about the wheel axis and by finding the intersection of rotating points and a perpendicular plane, the perpendicular plane is defined as any arbitrary plane which passes through the wheel axis. This method is useful particularly for obtaining geometrical parameters of a wheel such as flange height, flange slope and flange thickness. In order to commission the proposed method, a prototype system was designed and manufactured. The performance of the system, evaluated in different circumstances, shows a measurement error of up to 2%. Compared with classical methods utilizing a caliper or those which use expensive equipment or additional parts such as reference guides, the proposed method is easy to use and flexible. Also, a novel calibration method is utilized to calibrate the system accurately and freely.

show abstract

Section: Calibration Of the Dual-parallel Laser Beam Profilermentioning

confidence: 99%

A new flexible laser beam profiler for the inspection of train wheels

Torabi

Younesian

2020

Proceedings of the Institution of Mechanical Engineers, Part F:

View full text Add to dashboard Cite

show abstract

“…In the first stage, we only change the bias b k of the line in (8), and (x 0,k, t 0,k ) of the hyperbola in (11) or t 0,k of the hyperbola in (12) until there is no significant change in the output error. In the second stage, we adjust all parameters of line in (8) and hyperbola in (11) or (12).…”

Section: B Learning Rule For Hyperbolamentioning

confidence: 99%

“…Initially set up random parameter vectors. Then input data and adjust the parameter vector as (8) and (11) …”

Section: B Learning Rule For Hyperbolamentioning

confidence: 99%

“…Neural network had been developed to solve the HT problem [8]- [10]. The Hough transform neural network (HTNN) was designed for detecting lines, circles, and ellipses [8]- [10].…”

Section: Introductionmentioning

confidence: 99%

“…The Hough transform neural network (HTNN) was designed for detecting lines, circles, and ellipses [8]- [10]. But there is no application to hyperbola detection.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Hough Transform Neural Network for Seismic Pattern Detection

Huang¹,

You²,

Chen³

et al.

The 2006 IEEE International Joint Conference on Neural Network Proceedings

View full text Add to dashboard Cite

Hough transform neural network is adopted to detect line pattern of direct wave and hyperbola pattern of reflection wave in a seismogram. The distance calculation from point to hyperbola is calculated from the time difference. This calculation makes the parameter learning feasible. The neural network can calculate the total error for distance from point to patterns. The parameter learning rule is derived by gradient descent method to minimize the total error. Experimental results show that line and hyperbola can be detected in both simulated and real seismic data. The network can get a fast convergence. The detection results can automatically provide a reference and improve seismic interpretation.

show abstract

Hough Transform Neural Network for Seismic Pattern Detection

Huang

You

Chen

et al. 2006

Neural Information Processing

View full text Add to dashboard Cite

Abstract. Hough transform neural network is adopted to detect line pattern of direct wave and hyperbola pattern of reflection wave in a seismogram. The distance calculation from point to hyperbola is calculated from the time difference. This calculation makes the parameter learning feasible. The neural network can calculate the total error for distance from point to patterns. The parameter learning rule is derived by gradient descent method to minimize the total error. Experimental results show that line and hyperbola can be detected in both simulated and real seismic data. The network can get a fast convergence. The detection results can improve the seismic interpretation.

show abstract

A Hough transform system based on neural networks

Cited by 16 publications

References 17 publications

A new flexible laser beam profiler for the inspection of train wheels

A new flexible laser beam profiler for the inspection of train wheels

Hough Transform Neural Network for Seismic Pattern Detection

Hough Transform Neural Network for Seismic Pattern Detection

Contact Info

Product

Resources

About