国际群体智能与进化计算杂志

抽象的

Statistical Modeling for Image Fusion

Mojdeh Sohrabi

The purpose of image fusion is to mix information from multiple images of an equivalent scene into one image that ideally contains all the important features from each of the original images. The resulting fused image are going to be thus more suitable for human and machine perception or for further image processing tasks. Many image fusion schemes are developed within the past. In general, these schemes are often roughly classified into pixel-based and regionbased methods. It has been shown that comparable results are often achieved using both sorts of methods with added advantages for the region based approaches, mostly in terms of the likelihood of implementing more intelligent fusion rules. On the opposite hand, pixel based algorithms are simpler and thus easier to implement.