A New Approach for Segmentation and Classification of Textures by using Square Primitive Pattern

The Texture classification is an important task in the field of object recognition. The texture classification is highly dependent on the segmentation of objects in the input image. So, the present paper proposes a morphological approach to detect the objects and extract the edges of the detected objects. Then the present paper measures recurrence of the square primitive pattern (SPP) in each of the 3×3 local window. Based on this measurement, the textures are precisely classified. This is proved by the experimental results conducted on different textures.