Comparison of the Performance of Resnet-50 and Shufflenet Deep Learning Algorithms in Classification of Hazelnut Fruit
Published in BZT Turan, 2025
This study proposes an automatic classification system for hazelnut fruit using deep learning, comparing the performance of ResNet-50 and ShuffleNet algorithms on a dataset of 15,770 hazelnut images. ShuffleNet, with its lower resource demands, is recommended for real-time analysis due to its comparable accuracy (99.94% in testing) to ResNet-50 (99.97% in testing).