Classification of banana leaf diseases using enhanced gabor feature descriptor

Abstract

Banana leaf disease classification requires the application of machine learning and image processing on the images obtained from the field. A lot of diseases like banana black sigakota (BBS), banana bunchy top virus (BBTV), and banana bacterial wilt (BBW) have been reported on banana plantation. In this paper, the merits of both Gabor filter and 2D log Gabor filter have been used to construct an enhanced Gabor filter, which extracts features from the images of diseased plant. Further, KNN classifier is used in the classification of diseases. Classification results with diseased datasets show that the proposed method achieves better when compared to SIFT and SURF feature descriptors.