The objective of image fusion is to combine the source images of the same scene to form one composite image that contains a more accurate description of the scene than any one of the individual source images. A comparison of various feature based fusion schemes is presented in this study. Feature extraction plays a major a role in the implementation of feature-level fusion approaches. Prior to the merging of images, salient features, present in all source images, are extracted using an appropriate feature extraction procedure. Then, fusion is performed using these extracted features. The performance of image fusion is evaluated by normalized least square error, entropy, overall cross entropy, standard deviation and mutual information. The experimental results show that the images fused with salience match measure, gradient and gradient match measure gives better performance.