A novel cognitive transformation algorithm based on Gaussian cloud model and its application in image segmentation (Q1689456)

In the paper, the authors deal with the problem of transforming intension and extension using a cognitive transformation. As transformation they use a Gaussian cloud model (GCM). At the beginning, the authors give the definition of GCM together with its properties. Next, they give two algorithms that use GCM, namely the forward Gaussian cloud transformation (FGCT) and the backward Gaussian cloud transformation (BGCT). Moreover, the errors of the existing BGCT algorithms are analyzed in detail. Then, the authors introduce a new BGCT algorithm that is a multi-step one. Using this algorithm, they are able to obtain a stable concept intension in the cognitive process from extension to intension. It is obtained through the extraction of concepts' intension from the sample data. To show the effectiveness of the proposed method the authors make some experiments. In the first group of experiments they compared the proposed method with methods from the literature using the average absolute error and mean-squared error. Whereas in the second group they used the proposed algorithm in the image segmentation task and compared it with some of the existing segmentation methods, e.g. the type-2 fuzzy set method and the Otsu method.