Implementation of Synthetic Image Generation and Augmentation Using Conditional GANs for Rice Disease Identification and Expert System Design

Rice is one of the most important staple foods in the world. It plays a significant role in global food security. Crop productivity and quality are severely affected by various rice diseases. Traditional disease identification methods are time-consuming and expert-dependent. This conference paper presents the outcome of implementing a Conditional Generative Adversarial Network (cGAN)-based framework for synthetic image generation and augmentation to improve rice disease classification performance. The implemented methodology integrates synthetic image generation, hybrid dataset augmentation, and deep learning-based disease classification. A novel Convolutional Termite Life Cycle Network (CTLCN) architecture is introduced to optimize feature extraction and disease prediction.