Enhancing Plant Disease Identification through Innovative Image Synthesis: A Study of InstaGAN and Data Augmentation in Agriculture

This study explores the growing implementation of artificial intelligence (AI) technology, particularly in agriculture, where AI-driven object detection enhances efficiency. Deep learning's potential for plant disease identification is hampered by data scarcity. To address this, data augmentation techniques like Generative Adversarial Networks (GANs) are employed, with models like InstaGAN contributing to image synthesis. InstaGAN's innovative approach, integrating instance information and contextual preservation, is examined using the "Plant Village" dataset. The study underscores the value of InstaGAN in mitigating data limitations through synthesized image augmentation, showcasing promising outcomes in comparison to alternative models.