Defect Detection of Fruit Based on Deep Learning Approaches

Amirreza Rouhbakhsh Meghrazi¹ Shayan Nalbandian²

1) Department of Electronic Information, Northwestern Polytechnical University, Xi’An, Shaanxi, China, Email:
2) Department of Software, Northwestern Polytechnical University, Xi’An, Shaanxi, China,

Publication : 2nd International Congress on Science, engineering & New Technologies(secongress.com)
Abstract :
Fruits have a relatively short lifespan and are one of the most important sources of nutrition for humans. Fruit deterioration may happen at a number of times, including during harvest, transit, storage, etc. One criterion for determining the fruit s quality is its freshness. Before being consumed by people, around 25% of the produced fruits get spoiled for a variety of reasons. The rotting of one fruit directly affects the nearby ones. It is also one of the markers that estimate how long a fruit may be kept in storage. It is easier to remove rotten fruits from the whole lot by taking the proper action as soon as the deterioration is identified. In order to aid in keeping the fruits next to it from being spoiled. Recent advancements in technology, specifically in the field of deep learning, have greatly aided in the automated detection of damaged fruit. However, it is important to note that current methods only consider the surface characteristics of the fruit, disregarding any internal factors that may contribute to spoiling. To assess the freshness of fruits, a supervised learning approach is used with bananas being the chosen subject of this study. The researchers utilized a detection method, allowing for reliable differentiation between rotten and high-quality fruit.
Keywords : Machine Learning AI Deep Learning Defect Detection of Fruit Supervised Learning