### Abstract

We propose a new shape from texture method based on the ridge of continuous wavelet transform. This method determines the orientations of planar surface in a direct way under the perspective projection model. The variations of the image projected from a planar surface can be accurately characterized by the ridge of continuous wavelet transform. The ridge of 1-D signal and 2-D image are represented as ridge curve and ridge plane, respectively. Ridges represent the energy concentration in the time-frequency plane where the energy is a local maxima. We show that the ridge of the projected image is a parabolic plane with a rotation angle equal to the tilt angle of the planar surface. The ridge is then rotated with the angle such that the slant effect appears in the X-axis and pay no role along the Y-axis. As a result, the rotated ridge plane can be regarded as the plane composed of many 1-D ridge curve. The slant angle of 2-D image is thus obtained from the derived slant angle of 1-D signal. A voting method and a curve fitting method are developed to obtain the slant angle of 1-D signal. Several synthetic and real-world images have demonstrated the robustness and accuracy of our method.

Original language | English |
---|---|

Pages | 295-298 |

Number of pages | 4 |

Publication status | Published - 1996 Dec 1 |

Event | Proceedings of the 1996 IEEE International Conference on Image Processing, ICIP'96. Part 2 (of 3) - Lausanne, Switz Duration: 1996 Sep 16 → 1996 Sep 19 |

### Other

Other | Proceedings of the 1996 IEEE International Conference on Image Processing, ICIP'96. Part 2 (of 3) |
---|---|

City | Lausanne, Switz |

Period | 96-09-16 → 96-09-19 |

### Fingerprint

### All Science Journal Classification (ASJC) codes

- Hardware and Architecture
- Computer Vision and Pattern Recognition
- Electrical and Electronic Engineering

### Cite this

*Shape from texture based on the ridge of continuous wavelet transform*. 295-298. Paper presented at Proceedings of the 1996 IEEE International Conference on Image Processing, ICIP'96. Part 2 (of 3), Lausanne, Switz, .

}

**Shape from texture based on the ridge of continuous wavelet transform.** / Lu, Chun Shien; Hwang, Wen Liang; Liao, Hong Yung Mark; Chung, Pau-Choo.

Research output: Contribution to conference › Paper

TY - CONF

T1 - Shape from texture based on the ridge of continuous wavelet transform

AU - Lu, Chun Shien

AU - Hwang, Wen Liang

AU - Liao, Hong Yung Mark

AU - Chung, Pau-Choo

PY - 1996/12/1

Y1 - 1996/12/1

N2 - We propose a new shape from texture method based on the ridge of continuous wavelet transform. This method determines the orientations of planar surface in a direct way under the perspective projection model. The variations of the image projected from a planar surface can be accurately characterized by the ridge of continuous wavelet transform. The ridge of 1-D signal and 2-D image are represented as ridge curve and ridge plane, respectively. Ridges represent the energy concentration in the time-frequency plane where the energy is a local maxima. We show that the ridge of the projected image is a parabolic plane with a rotation angle equal to the tilt angle of the planar surface. The ridge is then rotated with the angle such that the slant effect appears in the X-axis and pay no role along the Y-axis. As a result, the rotated ridge plane can be regarded as the plane composed of many 1-D ridge curve. The slant angle of 2-D image is thus obtained from the derived slant angle of 1-D signal. A voting method and a curve fitting method are developed to obtain the slant angle of 1-D signal. Several synthetic and real-world images have demonstrated the robustness and accuracy of our method.

AB - We propose a new shape from texture method based on the ridge of continuous wavelet transform. This method determines the orientations of planar surface in a direct way under the perspective projection model. The variations of the image projected from a planar surface can be accurately characterized by the ridge of continuous wavelet transform. The ridge of 1-D signal and 2-D image are represented as ridge curve and ridge plane, respectively. Ridges represent the energy concentration in the time-frequency plane where the energy is a local maxima. We show that the ridge of the projected image is a parabolic plane with a rotation angle equal to the tilt angle of the planar surface. The ridge is then rotated with the angle such that the slant effect appears in the X-axis and pay no role along the Y-axis. As a result, the rotated ridge plane can be regarded as the plane composed of many 1-D ridge curve. The slant angle of 2-D image is thus obtained from the derived slant angle of 1-D signal. A voting method and a curve fitting method are developed to obtain the slant angle of 1-D signal. Several synthetic and real-world images have demonstrated the robustness and accuracy of our method.

UR - http://www.scopus.com/inward/record.url?scp=0030406452&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030406452&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0030406452

SP - 295

EP - 298

ER -