Leveraging Potential of Deep Learning for Fruit Quality Detection: A Review

Detection of the quality of fruits is essential for the health of humans as well as animals for sustainable agriculture. The quality of the fruit determines its price and directly impacts the agricultural sector thereby the economy. The early detection of the quality of the fruit can lead to the adoption of various techniques to further improve fruit quality, quantity and yield. Conventional methods include humans visually observing the fruit and checking for deformities. Moreover, IoT-based systems are used for this purpose,but satisfactory results have not been observed. Recent progress and research in deep learning methodologies have shown remarkable results for the aforementioned task which may lead to improvement in agricultural outcomes. However, the potential of deep learning methods for real-time fruit quality detection remains under-explored. This study aims to present a comprehensive review of widely adapted methods for fruit quality detection and provide future research scope and directions to the researchers as well as provide a clearer picture of quality determination using cutting-edge technologies.