Detection of various diseases in fruit and vegetables using deep learning techniques

Fruit and vegetable diseases have critical importance in terms of food safety and sustainable agricultural practices. Therefore, diseases reduce crop yields, reduce quality and thus threaten the global food supply. These diseases also negatively affect biodiversity, disrupt ecosystem balance and weaken farmers' livelihoods. In this study, the diseases seen in fruits and vegetables were determined by using deep learning techniques.Within the scope of this research, an online data set was obtained from 2907 RGB images belonging to 12 classes. The data set was increased from 2907 to 17442 with the data augumentıon method for each class.For the detection of various diseases in fruits and vegetables, a 10-layer convolutional deep network model was created and pre education deep network architectures (InceptionV3 and ResNet50) were employed. The obtained results were compared in terms of time and success rate to determine the most successful methods. The results of the analyzes provided were also carried out with this designed real-time system to detect disease images in fruits and vegetables and transfer their predictions to the computer screen. Keywords