Current Proceedings on Technology

Yazarlar: Saleh A. Alshehri

Konular:-

Anahtar Kelimeler:Image binarization,Image thresholding,Pixel classification.

Özet: Image binarization is an important step in many applications. With the existence of new imaging devices such as smart phones, the types of application become wider. Image binarization is the process of finding the proper image pixels threshold. Threshold methods can be grouped into two main categories: global and local. Local thresholding produces good binary images but is time consuming, which may not be appropriate for some imaging applications. On the other hand, global thresholding is fast but does not produce the target accuracy. In this article, a simple yet efficient method is presented. It is a mixture of global and local thresholding. It can be classified as local thresholding. However, it does not require calculations based on each pixel in the given image. It also automates the choice of thresholding free parameter. The resulting performance is comparable to the best of the existing methods and is faster than many of them.

---