Hough-transform detection of lines in 3-D space

Bhattacharya, Prabir; Liu, Haiying; Rosenfeld, Azriel; Thompson, Scott F.

doi:10.1016/s0167-8655(00)00044-1

Cited by 24 publications

(10 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For convenience of calculation, both Bhattachary et al 7 and Moqiseh and Nayebi 5 presented a combination of two 2-D HTs for the detection of a straight line in 3-D space. There are also some difficulties with these parameter space decomposition techniques.…”

Section: Three-dimensional Htmentioning

confidence: 99%

“…Bhattachary et al 7 used a priori knowledge of the number of peaks in the first Hough space, but when the number of targets is unknown and the observation time is short, it is more difficult to determine the peaks in the accumulator of the first HT. Figure 3(b) shows the accumulator obtained by applying the first HT to the accumulated image in Fig.…”

Section: Three-dimensional Htmentioning

confidence: 99%

“…0091-3286/2010/$25.00 C 2010 SPIE The classical method of reducing the computational cost is the parameter space decomposition technique, such as two 2-D HTs instead of a 3-D one to detect 3-D linear trajectories. [5][6][7] However, there are some difficulties with this parameter space decomposition technique. A generic disadvantage of such methods is that multistage process will cause systematic errors that propagate through the various stages.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Hough transform relative to a four-dimensional parameter space for the detection of constant velocity target

2010

Opt. Eng

View full text Add to dashboard Cite

We present a novel Hough transform method for moving point target detection by using a 4-D parameter space. A new representation, which uses four parameters (the distance variable ρ, the angle variable θ, the velocity variable v , and the distance variable S), is proposed for constant velocity target in the 3-D observation space X-Y-T. By estimating velocity, a target trajectory can be transformed into a 4-D parameter space with a limited range of projection options. Our simulation and analysis show that the new algorithm can produce positive results in suppressing noise points with less computational cost. C 2010 Society of Photo-Optical Instrumentation Engineers. IntroductionLong-range dim-point target detection is a key technology in many detection systems, and several methods can accomplish moving point target (MPT) detection and trajectory estimation with varying degrees of success. In a remote sensing image, a target is of the size of a pixel or smaller, for the reason that when the gray value of a point target is not the greatest one in the image, it is hard to locate the point target using a single threshold. Generally, the characteristics of motion are used for MPT detection for noise suppression under a low signal-to-noise ratio (SNR) condition. Since the longrange motion is usually continuous and regular in time and space, methods based on the time information are effective ways to solve the problem.In 1977, Falocner 1 used the Hough transform (HT) to detect dim targets by extracting the linear motion trajectory in two-dimensional (2-D) plane. From then on, more and more algorithms based on the HT have been proposed to detect MPT. Some algorithms 1, 2 applied a 2-D HT to find moving targets, since the projection of a constant velocity (CV) point target trajectory to the image plane can be modeled as a straight line in X-Y Cartesian coordinates. But a standard HT 3 will give many false trajectories due to a large search region and too much noise, 4 and the endpoints and connectivity of the target trajectory can not be easily determined or guaranteed, and all that may impact on the track prediction, especially in a limited observing time. Thus, the detection of an MPT is always performed in a three-dimensional (3-D) observation space, which consists of two spatial dimensions and one temporal dimension.The principal problem for 3-D HT is that this approach is impractical because the required storage is too large and the computational time is too long. To overcome this problem, researches have proposed many efficient implementations of the HT.

show abstract

Section: Three-dimensional Htmentioning

confidence: 99%

Section: Three-dimensional Htmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Hough transform relative to a four-dimensional parameter space for the detection of constant velocity target

2010

Opt. Eng

View full text Add to dashboard Cite

show abstract

“…In this section, the Hough transform is expanded for line detection in this 3-D space. A unique line in 3-dimensioanl space can be parameterized by 5 parameters: (φ, θ) it's two direction angles in spherical form, and (x1,y1,z1) the Cartesian coordinates of its intersection with a plane through the origin perpendicular to it [6]. (4) and (x, y, z) are the co-ordinates of any point on the line.…”

Section: -D Hough Transformmentioning

confidence: 99%

3-D Hough transform for surveillance radar target detection

Moqiseh

Nayebi

2008

2008 IEEE Radar Conference

View full text Add to dashboard Cite

A 3-D Hough transform is introduced for detection of linear trajectories in a 3-D data space. Data space is constructed from returned radar data using range, bearing, and time information. This information is then mapped into a 3-D Cartesian data space. The constant velocity target trajectories appear as straight lines in this space, and are detected by the usage of 3-D Hough transform. The efficacy of the 3-D Hough transform is verified by simulation. In comparison with the 2-D case, when only the range-time information is used, the 3-D Hough transform has a better performance.

show abstract

“…24 In a later work, straight lines in 3-D space have been detected as depth edges by using a (2 þ 2)-D Hough space. 25 Wani and Batchelor 26 have studied spatial curves to design edge masks for edge detection. This methodology has been further investigated based on the scan-line approximation technique.…”

Section: Introductionmentioning

confidence: 99%