圖像分割中的閥值選取是一個相當重要的技術，也延伸到許多領域當中，如特徵辨識、生物醫學影像等等，而選取的方法包含許多，P參數法、最大熵閥值法、Otsu法等都是可以作為選取閥值的方法。閥值選取基本上是一個像素分佈的問題，基於可以依像素特性將圖像分為兩類:一個是屬於目標部份，另一個則是背景部分，判斷的依據則為像素的灰階數值小於或等於閥值分為一類，像素的灰階數值大於閥值分為一類，此一技術已被提出且廣泛使用。針對Otsu法的延伸-二維的Otsu法(2D-Otsu)可以在當灰階直方圖並未具有雙峰值的特性存在條件下，得到一個較好的分割閥值，將目標及背景區隔開來，但須經過較複雜的計算過程。而PSO演算法是一個人工智慧演算法，具有參數少，收斂速度快等特性，可以成功地結合二維的Otsu法加速搜尋圖像分割閥值。Threshold selecting is a significant technique for image segmentation, which is applied broadly in many fields such as character recognition, analysis of biologic images etc. The method mainly includes P-tile method , the maximum entropy method , Otsu and so on. It is essentially a pixels classification problem. Its basic objective is to classify the pixels of a given image into two classes: one is those pertaining to an object and another is those pertaining to the background. While one includes pixels with gray values that are below or equal to a certain threshold, the other includes those with gray values above the threshold. As an extension of Otsu algorithm, two-dimensional Otsu algorithm (2D-Otsu) can give good result for those objects whose histogram does not have two peaks which represent objects and background, however, it costs complex computation. Particle swarm optimization (PSO) is a swarm intelligence optimization algorithm as a set few parameters, better global search capability, search results more stable and widely used. So we combined successfully these two algorithms to get ideal segmentation result with lesscomputation cost.
