The critical analysis of various image fusion techniques for enhanced image features interpretation in remote sensing applications

摘要

Multi sensor data fusion technique combines data and information from multiple sensors to achieve improved accuracies and better inference about the environment than single sensor. The paper presents an objective evaluation of three image fusion techniques. The fusions techniques based on Brovey Transform, Integration of substitution (IHS), discrete wavelet transform (DWT) using additive Wavelet (WT) was performed. The fused image is evaluated in 1:50000 scale. From visual aspect, the spatial and spectral resolutions of all the images have been enhanced compared with the source MS images. The result of fusion using IHS (both cylindrical and triangular models) fusion has more spatial resolution as compared to the other fusion methods although they exhibit color distortion for vegetation cover. The DWT fusion gave the optimum spectral enhancement when the level used was three. At higher levels, the color fades gradually. The IHS with DWT causes color distortion in the fused image, whereas the additive wavelet based fusion method preserves the original spectral content.