Analisa dan Perbandingan Algoritma Otsu Thresholding dengan Algoritma Region Growing Pada Segmentasi Citra Digital

Abstract

Segmentation is an important process in image processing which aims to split an image into several segments with certain criteria. Image segmentation is the separation of one object with other objects in an image or between objects with a background contained in an image. With this segmentation process, each object in the image can be taken individually so that it can be used as input. The purpose of image segmentation is to separate the object area from other regions based on similarity of image attributes so that it is easy to analyze in order to recognize objects for a purpose. This type of operation is closely related to pattern recognition. In general, the segmentation process is divided into three parts based on classification, by edge and by region. The segmentation process in digital images is done with edge detection to provide a target path along the edge (edge) of the image object. The algorithm used is the otsu thresholding algorithm and the region growing algorithm where the otsu thresholding algorithm can divide the gray image histogram into two different regions automatically. While the region growing algorithm is a procedure in which a set of pixels or subregions to a larger region is based on defined criteria.