本研究目的為改進由Avidan與Shamir所提出的影像細縫裁減（seam carving）演算法，seam carving是依據影像內容來對影像重新調整大小的演算法，被認為是一個有效解決影像縮放的方法，利用簡單的濾波器來找出影像中欲保留或移除之區塊，但是很多時候這種演算法並不能產生令人滿意地結果，它無法應付各種類型的影像，例如有複雜背景或是高反差顏色的影像。本研究方法：第一步驟、利用中間值濾波器來降低複雜背景的高頻雜訊；第二步驟、對影像進行邊緣偵測來計算能量值；第三步驟、使用分群法將影像前景和背景之能量差異強化。本研究結果：對於有複雜背景之影像利用本方法的三個步驟調整過後，皆能夠呈現出不錯的結果，如內文之影像。本研究結論：相信在未來只要有螢幕的裝置都能夠上網的話，在處理影像縮放的技術將會越顯重要，本研究方法將有助於未來顯示器的影像縮放，透過智慧型手機及各種多媒體裝置將運動資訊傳遞到使用者身上，讓相關運動知識與訊息傳遞更便利性，達成運動與生活密不可分，相信對推廣運動及相關產業必有相輔相成的作用。
The purpose of this study was to improve seam carving algorithm proposed by Avidan and Shamir. Seam carving is a content-aware image retargeting method. In particular, seam carving has gained attention as an effective solution, using simple filters to detect and preserve the important areas of an image. But in some cases this algorithm may not provide satisfactory results, especially for images having complex background or high-contrast colors. The purposed method has three steps, in the first step median filtering is performed on the given image to slightly reduce energy of complexity background. Then energy values are calculated by an edge detector. Finally, the difference between energy of the foreground and energy of the background is enhanced by clustering. The experimental results showed that the proposed method provides good results even for the images with complex background. This approach will help displays integrated with image resizing technology. Through smart phones and various multimedia devices, sports information and sports knowledge can be delivered to the user easily. Thus, not only will our life and sports become inextricable, but also the sports and related industries will be promoted.
