Review: Detection and Classification of Plants Leaf Diseases

Abstract

The aim of this project is to
design, implement and evaluate an image
processing software based solution for
automatic detection and classification of
plant leaf disease. We present fast,
automatic, cheap and accurate image
processing based solution. This solution is
composed of four main phases. First the
digital images are acquired from the field or
environment using digital camera. Image
preprocessing technique noise is removed
by filtering technique. Next, in the second
phase, the images are segmented using the
K-means clustering technique. In the third
phase, we calculate the texture features for
the segmented infected objects. Finally, in
the fourth phase the extracted features are
passed through a pre-trained neural
network