Design and Development of A Model to Classify Crop Foliar Diseases
Loading...
Date
item.page.authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
a controlled environment is a data item (image) that comprises only a single subject (leaf) and
newlinea white background. In an uncontrolled environment, an image contains the Mung leaf,
newlinebackground noise like stems, ground, other Mung leaves, etc. In combined environment images
newlineof both the controlled and uncontrolled environments are merged together. Seven different
newlineclassifiers namely Support Vector Machine (SVM), KNN (K Nearest Neighbor), AdaBoost
newline(Adaptive Boosting), GaussianNB (Gaussian Naive Bayes), DTC (Decision Tree Classifier),
newlineLogisticRegression and Custom CNNs with different architectures have been trained and
newlinecompared to each other.
newlineResearcher aims at detecting 3 mung leaf disease categories and a healthy leaf category.
newlineThe model extracts complex features of various diseases. Early detection will help farmers to
newlineimprove their productivity. The main objective was to automate Mung Leaf disease
newlineidentification using advanced machine learning and deep learning approaches and image data.
newlineAmong all the classifiers the custom CNN achieved performs well and achieved highest
newlineaccuracy in all the three environments. Custom CNN achieves 99.24% of training and 95.05%
newlineof testing accuracy in controlled environment. In uncontrolled environment custom CNN
newlineachieves 99.69% training and 87.88% of testing accuracy. In combined environment custom
newlineCNN achieves 98.81% of training and 90.68% of testing accuracy. The results shows high
newlinepotentiality of machine vision for recognition of diseased leaves. An interface is developed
newlinewhere user can input and image. Here are user can select image from either single leaf
newline(controlled environment), photo captured from the field itself (uncontrolled environment).
newlineImage given as input by interface will be given to model for classification whether it is healthy
newlineor having disease and if it is affected by disease then which disease the leaf has i.e. amongst
newlinethe three Cercospora Leaf Spot, Powdery Mildew, and Yellow Mosaic Virus. Interface is just a medium to interact with model, and model works as an