College of Computer Science and Information Technology, University of Kerbala has discussed the M.A thesis which is entitled ” Plant Diseases Classification Based On Leaves Using Deep Learning”.
Being submitted by Zahraa Yassin Younis, the study aims at developing an effective and accurate model to detect and diagnose diseases of tomato and potato plants using Deep Learning and Machine Learning techniques, while comparing the performance of different models to reach the best model in terms of accuracy and efficiency.
The study finds out that Densenet121 model outperformed other models in terms of accuracy, achieving 98.20% for potatoes and 95.44% for tomatoes.
The second proposal, where Densenet121 was used to extract features and SVM for classification, this approach shows higher performance with an accuracy of 99.75% for potatoes and 95.84% for tomatoes.
The study recommends using combined model (Densenet121 with SVM) to detect plant diseases due to its high accuracy and efficiency.
