影像融合技術是將多張相同視景,但是不同焦距之影像融合成一張新的影像,這張影像是由各個影像最清晰的部分組合而成。在一些圖形識別、醫學影像或物件辨識的應用上,有時我們無法只從一張影像獲得所有資訊,必須要藉著不同光源或焦距來取得多張影像甚至是連續影像,而必須經由融合,取出重要區域進行組合,資訊才得以完整呈現,在這篇論文提出一個方法,以頻率域融合再進行多階段空間域影像融合技術解決空間域及頻率域的問題,最後本研究將此方法以實際影像和模擬影像來做實驗,並且和其他影像融合的方法進行比較。
Image fusion technique combines multiple images of the same scene under variable foci. The result of image fusion is a new image which is composed of clear regions from those images. It is very useful in human and machine perception or further image-processing tasks such as pattern recognition, image segmentation, aid diagnosis and object recognition. In this paper, we proposed a novel method based on wavelet transform. We first extract DWTFM to fuse images for the initial step. With this composed image, multiple phases for further fusion are applied in the spatial domain. In the experiment, we apply this method both on synthesize and real images and compare to other fusion method.