應用影像融合技術於彩色影像之對比強化演算法

Abstract

由於攝影器材（例如數位相機和手機）的發展與普及，越來越多的數位影像出現在人們的日常生活中。同時，隨著網際網路與社群網路逐漸發展成熟，人們可以很輕易的與朋友分享彼此的影像。但是並不是所有的影像都是讓人滿意的。攝影器材的技術限制以及不適當的攝影環境會使得有些影像曝光不夠而有些影像則是過度曝光。為了能夠解決這個問題，許多傳統的影像強化技術被提出來。但是這些傳統技術經常只適用於一些特定的影像抑或這些方法是非自動化的。因此，本論文提供了一個基於影像融合技術的對比強化演算法。首先，數張亮度不同的影像會被產生出來。接著，輸入影像的像素會依據像素的亮度來做分群。最後，我們提出的Classified Image Fusion (CIF)方法會將這些虛擬影像作結合來得到一張曝光良好的結合後的影像。

There are more and more digital images in our daily life thanks to the popularity of photograph capturing equipments, such as digital cameras and mobile phones. In addition, as the Internet and social networks have been well developed, it’s easier for people to share images with their friends. However, not all people are satisfied with the photos they taken due to the limitations of the image capturing devices. The improper luminance condition may cause under-exposed and over-exposed images. To solve this problem, plenty of researches are proposed for contrast enhancement. However, they often cannot afford to produce pleasing images for a broad variety of low contrast images or cannot be automatically applied on all images. Hence, in this thesis, we propose a classified image fusion (CIF) method for image contrast enhancement. First several virtual images having different intensities are generated. Second, the input image pixels are classified to several classes according to their luminance values. Finally, CIF was proposed to combine these exposure images to produce a fused image in which every region is well-exposed.