The proposed scheme has been evaluated on Outex and CUReT databases, and the evaluation result shows that the proposed approach achieves an improved recognition rate compared to the previous research results.
In order to verify the effectiveness of the proposed SB-CLBP texture feature extraction method, a series of experiments have been conducted on two representative texture databases, i.e., the Outex database  and the CUReT database .
The Outex database, which includes the 24 classes of textures shown in Fig.
They are the Outex
database , Columbia-Utrecht Reflection and Texture (CUReT) database , UIUC database , and XU_HR database .
The collection is similar to Outex 0, except that this collection contains 2112 images (88 per texture) and that they are of size 64 x 64 pixels.
The collection is similar to Outex 0, except that this collection contains 8832 images (368 per texture) and that they are of size 32 x 32 pixels.
Table 5 indicates that the Extended ArTex achieves best results in domains with relatively large images (128 x 128 or more), and performs relatively poorly on domains with small images, like Outex 2 and Brodatz C.