Texture based classification of arecanut

Abstract

Crop grading is one of the important stages in crop management. The different grades can be done by classification. In this paper, we propose the texture based grading of arecanut. Different texture features are extracted from arecanut by applying approaches such as Wavelet, Gabor, Local Binary Pattern (LBP), Gray Level Difference Matrix (GLDM) and Gray Level Co-Occurrence Matrix (GLCM) features. For classification Nearest Neighbor (NN) classifier is used. Experimentation conducted using a dataset of 700 images of 7 classes to demonstrate the proposed model's performance. 91.43% of classification rate is achieved with Gabor wavelet features.