Gürültünün kenar belirleme tekniklerine etkisi

The subject of identification edge in images has a wide application in various fields for that it's considered one of the important topics in a digital image processing. There are many algorithms to detect the edge in images, but the performance of these algorithms depends on the type of image, the environment of the image and the threshold value of the edge algorithm. In this thesis five edge detection algorithms were evaluated by using several types of original images, these images were placed in multi-environments (clean, noisy, and de-noised). According to this evaluation results, the best edge detection algorithm and the best threshold value were found in each environment.