### Abstract

Diffusion and correlation effects are two principal phenomena which have been studied for years and several visualization techniques have been proposed to help scientists to understand them. The analysis of these phenomena will help to extract important information from data sets. To understand these problems we combine wavelet and entropy analysis to evaluate the evolution of these behaviors through scale and time. We present image case studies to show several different kinds of behaviors of these effects. Some of them are fallible cases and not reliable, as the images do not show the desired information. We calculate entropy of smooth and detail coefficient sets, generated by wavelet transform of these sample images in each scale, to obtain measures that allow us to evaluate these behaviors according to the organization complexity. These measures can provide an indication about the quality of the rendered images.

Original language | English (US) |
---|---|

Title of host publication | Proc 1997 10 Braz Symp Comput Graph Image Process SIBGRAPI 97 |

Editors | F. Henrique, M.L. Netto |

Publisher | IEEE Comp Soc |

Pages | 119-125 |

Number of pages | 7 |

State | Published - 1997 |

Event | Proceedings of the 1997 10th Brazilian Symposium of Computer Graphic and Image Processing, SIBGRAPI 97 - Campos do Jordao, Braz Duration: Oct 13 1997 → Oct 16 1997 |

### Other

Other | Proceedings of the 1997 10th Brazilian Symposium of Computer Graphic and Image Processing, SIBGRAPI 97 |
---|---|

City | Campos do Jordao, Braz |

Period | 10/13/97 → 10/16/97 |

### Fingerprint

### ASJC Scopus subject areas

- Engineering(all)

### Cite this

*Proc 1997 10 Braz Symp Comput Graph Image Process SIBGRAPI 97*(pp. 119-125). IEEE Comp Soc.

**Wavelet and entropy analysis combination to evaluate diffusion and correlation behaviors.** / Chiou, Rui C H; Ferreira, Maria Alice G V; Silva, Claudio T.; Kaufman, Arie E.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*Proc 1997 10 Braz Symp Comput Graph Image Process SIBGRAPI 97.*IEEE Comp Soc, pp. 119-125, Proceedings of the 1997 10th Brazilian Symposium of Computer Graphic and Image Processing, SIBGRAPI 97, Campos do Jordao, Braz, 10/13/97.

}

TY - GEN

T1 - Wavelet and entropy analysis combination to evaluate diffusion and correlation behaviors

AU - Chiou, Rui C H

AU - Ferreira, Maria Alice G V

AU - Silva, Claudio T.

AU - Kaufman, Arie E.

PY - 1997

Y1 - 1997

N2 - Diffusion and correlation effects are two principal phenomena which have been studied for years and several visualization techniques have been proposed to help scientists to understand them. The analysis of these phenomena will help to extract important information from data sets. To understand these problems we combine wavelet and entropy analysis to evaluate the evolution of these behaviors through scale and time. We present image case studies to show several different kinds of behaviors of these effects. Some of them are fallible cases and not reliable, as the images do not show the desired information. We calculate entropy of smooth and detail coefficient sets, generated by wavelet transform of these sample images in each scale, to obtain measures that allow us to evaluate these behaviors according to the organization complexity. These measures can provide an indication about the quality of the rendered images.

AB - Diffusion and correlation effects are two principal phenomena which have been studied for years and several visualization techniques have been proposed to help scientists to understand them. The analysis of these phenomena will help to extract important information from data sets. To understand these problems we combine wavelet and entropy analysis to evaluate the evolution of these behaviors through scale and time. We present image case studies to show several different kinds of behaviors of these effects. Some of them are fallible cases and not reliable, as the images do not show the desired information. We calculate entropy of smooth and detail coefficient sets, generated by wavelet transform of these sample images in each scale, to obtain measures that allow us to evaluate these behaviors according to the organization complexity. These measures can provide an indication about the quality of the rendered images.

UR - http://www.scopus.com/inward/record.url?scp=38449121331&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38449121331&partnerID=8YFLogxK

M3 - Conference contribution

SP - 119

EP - 125

BT - Proc 1997 10 Braz Symp Comput Graph Image Process SIBGRAPI 97

A2 - Henrique, F.

A2 - Netto, M.L.

PB - IEEE Comp Soc

ER -