A REVIEW OF IMAGE SEGMENTATION USING TWO LEVEL THRESHOLD AND FCM

ABSTRACT

Image linearization is the process of dividing pixel values into two groups, black for the background and white for the foreground. There are two forms of thresholding: global thresholding and local thresholding. This paper proposes a locally adaptive thresholding approach that eliminates background by using the local mean and standard deviation. Thresholding is the most common and simple way for segmenting an image. To evaluate the quality of a segmented image, statistical measurements such as the Jaccard Similarity Coefficient and the Peak Signal to Noise Ratio are utilized (PSNR).

FULL TEXT