Theoretical Study of Image Fusion Techniques: A Review

Abstract

the goal of image fusion is to combine relevant
information from two or more images of the same view in to
single image. The result of image fusion is a new fused image
which is more suitable for human being and machine
discernment for further image-processing tasks like
segmentation, feature taking out and objects recognition. In this
paper the image fusion techniques described using the PCA, and
wavelet family. Principal component analysis (PCA) is a wellknown scheme for feature extraction and dimension reduction.
In DCT low frequency region of the image has large DCT coefficient. So it has very good energy compactness properties. In
DWT image are di viding in low sub bands and high sub bands
are fused using various fusion methods. Finally, the output of the
fused image is obtained by applying inverse wavelet transform
on the fused coefficients of low sub bands and high sub bands.
Where in curvelet it given smooth cured edge detection. Above
technique mainly done in two domain: spatial domain and
transform domain where it performed fusion at three different
processing levels which are pixel level, feature level and decision
level according to the stage at which the fusion takes place. This
is depends on the required application