### Abstract

In this paper, we propose two new spatially adaptive image fusion algorithms based on Bayesian approach for merging remotely sensed panchromatic and multi-spectral images. The two complementary images are modeled as correlated two dimensional stochastic signals and the high-resolution multi-spectral image is estimated by minimizing the mean squared error between the original high-resolution image and the estimated image. We assume that the estimator is locally linear and obtain the local linear minimum mean square error (MMSE) estimator for image fusion. Two MMSE image fusion algorithms are derived on different assumptions of the images. If we assume that pixels in the images are uncorrelated with their neighbors, the estimator becomes a point processor which is controlled by an adaptive gain expressed by the ratio of local cross-covariance between the two images and the local variance of the panchromatic image. On the other hand, if we assume that pixels in a small block are considered stationary and correlated with one another, the estimator uses the locally stationary cross-covariance matrix between the two images and auto-covariance matrix of the panchromatic image. For the second algorithm, we take Fast Fourier Transform (FFT) based approach in order to avoid complex matrix computations and achieve a fast algorithm. Experimental results show that the proposed algorithms are superior to conventional algorithms according to visual and quantitative comparisons.

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

Title of host publication | Image Processing |

Subtitle of host publication | Algorithms and Systems, Neural Networks, and Machine Learning - Proceedings of SPIE-IS and T Electronic Imaging |

DOIs | |

Publication status | Published - 2006 Apr 17 |

Event | Image Processing: Algorithms and Systems, Neural Networks, and Machine Learning - San Jose, CA, United States Duration: 2006 Jan 16 → 2006 Jan 18 |

### Publication series

Name | Proceedings of SPIE - The International Society for Optical Engineering |
---|---|

Volume | 6064 |

ISSN (Print) | 0277-786X |

### Other

Other | Image Processing: Algorithms and Systems, Neural Networks, and Machine Learning |
---|---|

Country | United States |

City | San Jose, CA |

Period | 06/1/16 → 06/1/18 |

### Fingerprint

### All Science Journal Classification (ASJC) codes

- Electronic, Optical and Magnetic Materials
- Condensed Matter Physics
- Computer Science Applications
- Applied Mathematics
- Electrical and Electronic Engineering

### Cite this

*Image Processing: Algorithms and Systems, Neural Networks, and Machine Learning - Proceedings of SPIE-IS and T Electronic Imaging*[60640T] (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 6064). https://doi.org/10.1117/12.642765

}

*Image Processing: Algorithms and Systems, Neural Networks, and Machine Learning - Proceedings of SPIE-IS and T Electronic Imaging.*, 60640T, Proceedings of SPIE - The International Society for Optical Engineering, vol. 6064, Image Processing: Algorithms and Systems, Neural Networks, and Machine Learning, San Jose, CA, United States, 06/1/16. https://doi.org/10.1117/12.642765

**Spatially adaptive multi-resolution multi-spectral image fusion based on bayesian approach.** / Park, Jong Hyun; Kang, Moon Gi.

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

TY - GEN

T1 - Spatially adaptive multi-resolution multi-spectral image fusion based on bayesian approach

AU - Park, Jong Hyun

AU - Kang, Moon Gi

PY - 2006/4/17

Y1 - 2006/4/17

N2 - In this paper, we propose two new spatially adaptive image fusion algorithms based on Bayesian approach for merging remotely sensed panchromatic and multi-spectral images. The two complementary images are modeled as correlated two dimensional stochastic signals and the high-resolution multi-spectral image is estimated by minimizing the mean squared error between the original high-resolution image and the estimated image. We assume that the estimator is locally linear and obtain the local linear minimum mean square error (MMSE) estimator for image fusion. Two MMSE image fusion algorithms are derived on different assumptions of the images. If we assume that pixels in the images are uncorrelated with their neighbors, the estimator becomes a point processor which is controlled by an adaptive gain expressed by the ratio of local cross-covariance between the two images and the local variance of the panchromatic image. On the other hand, if we assume that pixels in a small block are considered stationary and correlated with one another, the estimator uses the locally stationary cross-covariance matrix between the two images and auto-covariance matrix of the panchromatic image. For the second algorithm, we take Fast Fourier Transform (FFT) based approach in order to avoid complex matrix computations and achieve a fast algorithm. Experimental results show that the proposed algorithms are superior to conventional algorithms according to visual and quantitative comparisons.

AB - In this paper, we propose two new spatially adaptive image fusion algorithms based on Bayesian approach for merging remotely sensed panchromatic and multi-spectral images. The two complementary images are modeled as correlated two dimensional stochastic signals and the high-resolution multi-spectral image is estimated by minimizing the mean squared error between the original high-resolution image and the estimated image. We assume that the estimator is locally linear and obtain the local linear minimum mean square error (MMSE) estimator for image fusion. Two MMSE image fusion algorithms are derived on different assumptions of the images. If we assume that pixels in the images are uncorrelated with their neighbors, the estimator becomes a point processor which is controlled by an adaptive gain expressed by the ratio of local cross-covariance between the two images and the local variance of the panchromatic image. On the other hand, if we assume that pixels in a small block are considered stationary and correlated with one another, the estimator uses the locally stationary cross-covariance matrix between the two images and auto-covariance matrix of the panchromatic image. For the second algorithm, we take Fast Fourier Transform (FFT) based approach in order to avoid complex matrix computations and achieve a fast algorithm. Experimental results show that the proposed algorithms are superior to conventional algorithms according to visual and quantitative comparisons.

UR - http://www.scopus.com/inward/record.url?scp=33645680846&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645680846&partnerID=8YFLogxK

U2 - 10.1117/12.642765

DO - 10.1117/12.642765

M3 - Conference contribution

AN - SCOPUS:33645680846

SN - 0819461040

SN - 9780819461049

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Image Processing

ER -