Cairo University, Faculty of Computers & Information Information Technology Department 5 Ahmed Zewal St., Orman, Giza, office box# 12613 (email)
The study described in this paper consists of a
method that applies Gabor wavelet transform technique to extract
relevant features related to image of tomato leaf in conjunction
with using Support Vector Machines (SVMs) with alternate
kernel functions in order to detect and identify type of disease
that infects tomato plant. Initially, we collected real samples of
diseased tomato leaves, next we isolated each leaf in single image,
wavelet based feature technique has been employed to identify
an optimal feature subset. Finally, a support vector machine
classifier with different kernel functions including Cauchy kernel,
Invmult Kernel and Laplacian Kernel was employed to evaluate
the ability of this approach to detect and identify where tomato
leaf infected with Powdery mildew or early blight. To evaluate the
performance of presented approach, we present tests on dataset
consisted of 100 images for each type of tomato diseases. Extensive
experimental results demonstrate that the proposed approach
provides excellent annotation with accuracy 99.5 %. Efficient
result obtained from the proposed approach can lead to tighter
connection between agriculture specialists and computer system,
yielding more effective and reliable results.