外文摘要：In many agricultural products, information technologies are utilized in classification processes at the desired quality. It is undesirable to mix different types of cherries, especially in export-type cherries. In this study on cherries, one of the important export products of Turkey, the classification of cherry species was carried out with ensemble learning methods. In this study, a new dataset consisting of 3570 images of seven different cherry species grown in Isparta region was created. The generated new dataset was trained with six different deep learning models with pre-learning on the original and incremental dataset. As a result of the training with incremental data, the best result was obtained from the DenseNet169 model with an accuracy of 99.57%. The two deep learning models with the best results were transferred to ensemble learning and a 100% accuracy rate was obtained with the Maximum Voting model.
外文关键词：deep learning;Classification;ensemble learning;CNN;Cherry species
作者：Kayaalp, Kiyas
作者单位：Isparta Univ Appl Sci
期刊名称：EUROPEAN FOOD RESEARCH AND TECHNOLOGY
期刊影响因子：0.0
出版年份：2024
出版刊次：250(5)
原文传递申请：江苏省科技资源（工程技术文献）统筹服务平台