FRUIT RIPENESS DETECTION USING DEEP LEARNING

Abstract—The agricultural industry has been facing challenges in traditional and manual visual grading of fruits due to its laborious nature inconsistent inspection classification process. To accurately estimate yield automate harvesting, it is crucial classify the based on their ripening stages. However, can be difficult differentiate between stages same fruit variety high similarity images during cycle. address these challenges, we plan develop an accurate, fast, reliable detection sys- tem using deep learning techniques. The modernization crops offers opportunities for better quality harvests significant cost savings. Our approach involves adapting state-of-the- art object detector faster R-CNN, transfer learning, detect from obtained through modalities such as colour (RGB) Hyper Spectral Imaging System (HSI). system’s robustness will enable us varieties determine stage a particular with effectiveness accuracy. system also efficiently segment multiple instances image grade individual objects. Index Terms—Modernisation, HSI, CNN, RGB