Derin öğrenme yöntemleri ile meyvelerin tazelik durumunun belirlenmesi
Yükleniyor...
Dosyalar
Tarih
2022
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Tekirdağ Namık Kemal Üniversitesi
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Bilindiği üzere günümüzde derin öğrenme teknolojisi yüz tanıma, ses tanıma, güvenlik sektörü, savunma sanayi gibi pek çok alanda kullanılmaktadır. Pek çok katmanlardan oluşan derin öğrenme, makine öğrenmesinin bir dalıdır. Derin öğrenme ile modele bir veri kümesi vererek çıktıların tahmin edilmesi sağlanabilir. Günümüzde meyve tazeliği önemli sağlık, çevre ve ekonomik sorunlardan biridir. Bu çalışmada seçilen üç meyve; elma, portakal ve muz üzerinde çalışmalar yapılmıştır. Çalışmada iki tane VGG16 baz alınarak model geliştirilmiş ve iki tane VGG19 baz alınarak model geliştirilmiştir. Bu geliştirilen modeller, elde bulunan üç meyve türünün veri setleri üzerinde test edilerek en iyi sonucu hangisinden alınabileceği saptanmış ve modeller birbiri ile kıyaslanmıştır.
As it is known, deep learning technology is used in many areas such as face recognition, voice recognition, the security sector, defense industry. Deep learning, which consists of many layers, is a branch of machine learning. With deep learning, it is possible to predict the outputs by giving a dataset to the model. Today, fruit freshness is one of the important health, environmental and economic problems. Three fruits were selected in this study; Studies have been done on apples, oranges, and bananas. In the study, two models were developed based on VGG16 and two models were developed based on VGG19. These developed models were tested on the data sets of the three fruit types at hand, and it was determined which one could get the best results, and the models were compared with each other.
As it is known, deep learning technology is used in many areas such as face recognition, voice recognition, the security sector, defense industry. Deep learning, which consists of many layers, is a branch of machine learning. With deep learning, it is possible to predict the outputs by giving a dataset to the model. Today, fruit freshness is one of the important health, environmental and economic problems. Three fruits were selected in this study; Studies have been done on apples, oranges, and bananas. In the study, two models were developed based on VGG16 and two models were developed based on VGG19. These developed models were tested on the data sets of the three fruit types at hand, and it was determined which one could get the best results, and the models were compared with each other.
Açıklama
Anahtar Kelimeler
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Computer Engineering and Computer Science and Control, Görüntü işleme-bilgisayarlı, Image processing-computer assisted
