Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Texture Classification Based on Cross and Diagonal Shape Descriptor Co-Occurrence Matrix


Affiliations
1 Dept of CSE, RGMCET, Nandyal, A.P., India
2 Dean of CSE Dept, AGI, Hyderabad, AP, India
3 Dept of CSE, JNTUA, Anantapuram, India
     

   Subscribe/Renew Journal


In the proposed method an image texture is divided in to two sets of images based on the cross texture unit elements (CTUE) and diagonal texture unit element (DTUE). The CTUE and DTUE represent four pixels. The CTUE and DTUE are represented as two separate 2 x 2 grids. Shape descriptors indexes (SDI) are evaluated on the 2 x 2 grids of CTUE and DTUE. Based on the SDI of CTUE and DTUE two different texton images are formed. On these two texton images, gray level co-occurrence matrix (GLCM) features are evaluated individually for a classification purpose. The proposed method is experimented on different textures. The results indicate the efficacy of proposed method over the other methods.


Keywords

Texture, Texture Unit Element, Shape Descriptor, GLCM.
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 213

PDF Views: 3




  • Texture Classification Based on Cross and Diagonal Shape Descriptor Co-Occurrence Matrix

Abstract Views: 213  |  PDF Views: 3

Authors

P. Kiran Kumar Reddy
Dept of CSE, RGMCET, Nandyal, A.P., India
V. Vijaya Kumar
Dean of CSE Dept, AGI, Hyderabad, AP, India
B. Eswara Reddy
Dept of CSE, JNTUA, Anantapuram, India

Abstract


In the proposed method an image texture is divided in to two sets of images based on the cross texture unit elements (CTUE) and diagonal texture unit element (DTUE). The CTUE and DTUE represent four pixels. The CTUE and DTUE are represented as two separate 2 x 2 grids. Shape descriptors indexes (SDI) are evaluated on the 2 x 2 grids of CTUE and DTUE. Based on the SDI of CTUE and DTUE two different texton images are formed. On these two texton images, gray level co-occurrence matrix (GLCM) features are evaluated individually for a classification purpose. The proposed method is experimented on different textures. The results indicate the efficacy of proposed method over the other methods.


Keywords


Texture, Texture Unit Element, Shape Descriptor, GLCM.