An automated approach for fruits and vegetable image enhancement and classification using GAN

Abstract Fruits and vegetables are essential components of human nutrition, providing vital vitamins, minerals, antioxidants, and dietary fibers that help prevent chronic diseases and maintain overall well-being. In modern society, the accurate recognition and classification of these food items have become increasingly important across multiple domains: agriculture, food processing, retail markets, health monitoring, and supply chain automation. Traditional recognition methods rely on manual inspection, which is highly subjective, labor-intensive, and prone to errors, particularly when dealing with large quantities of produce. With the rise of artificial intelligence (AI), particularly computer vision and deep learning, automated fruit and vegetable recognition systems have gained traction. However, a major challenge persists: image quality. In real-world environments, images of fruits and vegetables are often captured under poor lighting conditions, with occlusions, motion blur, or low resolution due to inexpensive cameras or mobile devices. Such degraded images significantly reduce classification accuracy. To address these limitations, this work proposes an automated approach for fruits and vegetable image enhancement and classification using Generative Adversarial Networks (GANs). By enhancing image quality before classification, the system improves recognition accuracy, making it robust and practical for real-world applications. Such as grading of fruits at market. the work is helpful for farmers and online shopping sites for quality check