Arithmetic and Frequency Filtering Methods of Pixel-Based Image Fusion Techniques
In remote sensing, image fusion technique is a useful tool used to fuse high spatial resolution panchromatic images (PAN) with lower spatial resolution multispectral images (MS) to create a high spatial resolution multispectral of image fusion (F) while preserving the spectral information in the multispectral image (MS).There are many PAN sharpening techniques or Pixel-Based image fusion techniques that have been developed to try to enhance the spatial resolution and the spectral property preservation of the MS. This paper attempts to undertake the study of image fusion, by using two types of pixel-based image fusion techniques i.e. Arithmetic Combination and Frequency Filtering Methods of Pixel-Based Image Fusion Techniques. The first type includes Brovey Transform (BT), Color Normalized Transformation (CN) and Multiplicative Method (MLT). The second type include High-Pass Filter Additive Method (HPFA), High-Frequency-Addition Method (HFA) High Frequency Modulation Method (HFM) and The Wavelet transform-based fusion method (WT). This paper also devotes to concentrate on the analytical techniques for evaluating the quality of image fusion (F) by using various methods including Standard Deviation (SD), Entropy(En), Correlation Coefficient (CC), Signal-to Noise Ratio (SNR), Normalization Root Mean Square Error (NRMSE) and Deviation Index (DI) to estimate the quality and degree of information improvement of a fused image quantitatively.