在現實生活中硬幣是不可獲缺的交易媒介，然而現行的硬幣辨識卻無法同時進行多枚硬幣的判別，因此能有效的同時進行多枚硬幣辨識必能提升人類追求自動化的速度。本論文目的在於應用影像處理技術於自動化硬幣檢測。本研究所提技術可以有效處理「同時多硬幣處理」、「複雜背景相容」、「部分重疊問題」等問題。首先利用影像前處理的步驟，將擷取的影像做色彩空間的轉換，接著應用邊緣偵測技術擷取錢幣的輪廓，降低複雜背景所產生的影響。再使用樣板比對的方法來偵測出每張影像上的多個錢幣。 In real life, coins are indispensable transaction media. However, the existing coin identification can not identify multiple coins synchronously. Therefore, the synchronous and effective multi-coin identification is bound to enhance people’s speed at the pursuit of automation. The purpose of this paper is to apply the image processing technology to automatic coin identification. The technology proposed in this paper can effectively process the problems such as “Synchronous Multi-coin Processing”, “Complex Background Compatibility”, and “Problem of Partial Overlapping”. First, the steps of image pre-processing are utilized to make the retrieved images converted, regarding colors and space. Next, the Edge-Detection Technique is utilized to retrieve the profiles of the coins, for reducing the influence caused by a complex background. Then, the Template Matching Method is utilized to detect multiple coins on each image.
