Plant disease detection identifies the abnormalities in the plants caused by any bacterial or fungus diseases. The main aim of plant disease detection is to prevent the disease from spreading to other plants and reduce maintenance costs. Traditional methods for plant disease detection are difficult to achieve better detection in the complex natural environment.

   Deep learning methods are effective models for plant disease detection in complex surroundings. The advantages of deep learning models for plant disease detection have improved accuracy, generality, and training efficiency. However, deep learning models need a large amount of data to create an effective model and avoid over-fitting problems. Lack of data is the main issue faced by deep learning models. Data augmentation tackle such issue by generating new data based on the existing data. Data augmentation improves the model’s precision accuracy and reduces the cost of collecting and labeling data. The performance of plant disease detection is improved while applying data augmentation techniques in deep learning models.