Review of Image Fusion Techniques

---------------------------------------------------------------------***--------------------------------------------------------------------Abstract Image fusion is a procedure of fusing two or more images of same scene to form single fused image which displays vital information in the fused image. Image fusion technique is used for removing noise from the images. Noise is an unwanted material which deteriorates the quality of an image affecting the clarity of an image. Noise can be of various types such as Gaussian noise, impulse noise, uniform noise etc. Images corrupt sometimes during acquisition or transmission or due to fault memory locations in the hardware. Image fusion can be done at three levels such as pixel level fusion, feature level fusion and decision level fusion. There are mainly two types of image fusion techniques which are spatial domain fusion techniques and temporal domain fusion techniques. Average method, PCA fusion, high pass filtering are spatial domain methods and methods which include transformation such as Discrete Cosine Transform, Discrete wavelet transform are temporal domain fusion methods. There are various methods of image fusion which have many advantages and disadvantages. Many techniques suffer from the problem of color artefacts that comes in the fused image formed. In this paper different techniques have been reviewed and the main objective is to find the shortcomings that occur in different image fusion techniques.