Multi-sensor fusion of Landsat 8 thermal infrared (TIR) and panchromatic (PAN) images.

Hyung-Sup Jung,Sung-Whan Park

doi:10.3390/s141224425

Abstract

Data fusion is defined as the combination of data from multiple sensors such that the resulting information is better than would be possible when the sensors are used individually. The multi-sensor fusion of panchromatic (PAN) and thermal infrared (TIR) images is a good example of this data fusion. While a PAN image has higher spatial resolution, a TIR one has lower spatial resolution. In this study, we have proposed an efficient method to fuse Landsat 8 PAN and TIR images using an optimal scaling factor in order to control the trade-off between the spatial details and the thermal information. We have compared the fused images created from different scaling factors and then tested the performance of the proposed method at urban and rural test areas. The test results show that the proposed method merges the spatial resolution of PAN image and the temperature information of TIR image efficiently. The proposed method may be applied to detect lava flows of volcanic activity, radioactive exposure of nuclear power plants, and surface temperature change with respect to land-use change.

Highlights

IntroductionOver the past two decades, much attention has been paid to multi-sensor data fusion for remote sensing applications such as Earth surface displacement measurements via the fusion of X-, C- and
Over the past two decades, much attention has been paid to multi-sensor data fusion for remote sensing applications such as Earth surface displacement measurements via the fusion of X, C- andL-band SAR images [1], image classification through the fusion of optic and SAR images [2,3], feature extraction with the fusion of Lidar data and optical image [4], soil moisture retrieval by the integrated use of MODIS and advanced microwave scanning radiometer (AMSR-E)data [5], vegetation monitoring from the fusion of optic and SAR images [6,7], etc
We have introduced a scaling factor to control the trade-off between the spatial details and the thermal information

Summary

Introduction

Over the past two decades, much attention has been paid to multi-sensor data fusion for remote sensing applications such as Earth surface displacement measurements via the fusion of X-, C- and. Data [5], vegetation monitoring from the fusion of optic and SAR images [6,7], etc. One of multi-sensor advantages is an improved observability. Broadening the baseline of physical observables can result in remarkable improvements [8,9]. The data fusion of panchromatic (PAN) and multi-spectral (MS) images has been widely used to create fused images with high spatial and spectral resolutions [10,11,12]. The data fusion of PAN and thermal infrared (TIR) images is a good example of improved observability.

Methods

Results

Conclusion