期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2015
卷号：73
期号：2
出版社：Journal of Theoretical and Applied
摘要：Texture analysis is considered fundamental and important in the fields of pattern recognition, computer vision and image processing. Texture analysis mainly aims to computationally represent an intuitive perception of texture and to facilitate automatic processing of the texture information for artificial vision systems. In this paper, we have compared between texture classification methods based on the Random Forest (RF) and Support Vector Machine (SVM) classifiers by using various extraction feature methods namely bi-orthogonal wavelet transform, gray level histogram and co-occurrence matrices. Each of these methods has used to classify the image separately at first, and they have combined together secondly. Experiments were conducted on two different databases. The first texture database is CUReT and the second database was collected from Outex database. The results have revealed that, RF and SVM have yielded higher classification precision.
关键词：Supervised Random Forest; Support Vector Machine; Feature Extraction; Texture classification
