Bark Classification Using RBPNN in Different Color Space

Abstract — This paper presents an algorithm for feature extraction from a bark image set based on parameter of generalized Gaussian density (GGD) model and color angles in different color space. The extracted features such as the scale parameter and the shape parameter of GGD for image will be employed for classification. The radial basis probabilistic neural networks (RBPNN) and supporting vector machine (SVM) were used as the classifiers for bark image classification. The experimental results about the data set with 300 bark images show that the proposed feature extraction algorithm is a promising technique.