A new Convolutional Neural Network proposal for classifying various images

Yeşim Tiraki; Hasan Temurtaş; Soydan Serttaş; Çiğdem Bakır

Makale detayları

A new Convolutional Neural Network proposal for classifying various images

Yazarlar:

Yeşim Tiraki | Hasan Temurtaş | Soydan Serttaş | Çiğdem Bakır

Yayıncı:

Nigde Omer Halisdemir University

Ana Başlıklar:

Derin Öğrenme

Dergi:

Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.13 Issue 1, pp.262-278

Belge Tipi:

Makale

Yayın Tarihi:

1/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001309732

pISSN:

2564-6605

Anahtar kelimeler:

Özet:

 Nowadays, deep learning methods are widely used in many fields such as robotics, sound processing, medicine and imagery. In this study, studies on image classification and analysis in the literature were examined in detail. In addition, performance analyzes of deep learning methods were carried out using CNN on 4 different data sets named MNIST, Fashion MNIST, CIFAR-10 and CIFAR-100. The structures of the created deep learning models, the parameter values used for training, the layers used, the complexity matrices obtained for the validation data, accuracy and loss graphs are shown in detail. Our study was carried out with a different network structure consisting of 3 convolution layers, 3 batch normalization layers, 2 maxpooling layers, 1 flatten, 2 droupout, 2 dense layers. In addition, the performance of image classification on different data has been increased with our proposed model. At the end of the test, the accuracy results of our study were compared according to various evaluation metrics. For all the data sets used, the best and worst images were determined. With the proposed CNN model, high accuracy rates were observed for MNIST and Fashion MNIST datasets, with these values being 99.22 and 99.21, respectively.

 Nowadays, deep learning methods are widely used in many fields such as robotics, sound processing, medicine and imagery. In this study, studies on image classification and analysis in the literature were examined in detail. In addition, performance analyzes of deep learning methods were carried out using CNN on 4 different data sets named MNIST, Fashion MNIST, CIFAR-10 and CIFAR-100. The structures of the created deep learning models, the parameter values used for training, the layers used, the complexity matrices obtained for the validation data, accuracy and loss graphs are shown in detail. Our study was carried out with a different network structure consisting of 3 convolution layers, 3 batch normalization layers, 2 maxpooling layers, 1 flatten, 2 droupout, 2 dense layers. In addition, the performance of image classification on different data has been increased with our proposed model. At the end of the test, the accuracy results of our study were compared according to various evaluation metrics. For all the data sets used, the best and worst images were determined. With the proposed CNN model, high accuracy rates were observed for MNIST and Fashion MNIST datasets, with these values being 99.22 and 99.21, respectively.

 Nowadays, deep learning methods are widely used in many fields such as robotics, sound processing, medicine and imagery. In this study, studies on image classification and analysis in the literature were examined in detail. In addition, performance analyzes of deep learning methods were carried out using CNN on 4 different data sets named MNIST, Fashion MNIST, CIFAR-10 and CIFAR-100. The structures of the created deep learning models, the parameter values used for training, the layers used, the complexity matrices obtained for the validation data, accuracy and loss graphs are shown in detail. Our study was carried out with a different network structure consisting of 3 convolution layers, 3 batch normalization layers, 2 maxpooling layers, 1 flatten, 2 droupout, 2 dense layers. In addition, the performance of image classification on different data has been increased with our proposed model. At the end of the test, the accuracy results of our study were compared according to various evaluation metrics. For all the data sets used, the best and worst images were determined. With the proposed CNN model, high accuracy rates were observed for MNIST and Fashion MNIST datasets, with these values being 99.22 and 99.21, respectively.

 Günümüzde derin öğrenme metotları robotik, ses işleme, tıp ve görüntü gibi birçok alanda yaygın kullanılmaktadır. Bu çalışmada literatürde yapılan görüntü sınıflandırma ve analizine yönelik çalışmalar detaylı bir şekilde incelenmiştir. Ayrıca yapılan çalışmada CNN kullanılarak derin öğrenme metotlarının MNIST, Fashion MNIST, CIFAR-10 ve CIFAR-100 ismindeki 4 farklı veri seti üzerinde performans analizleri yapılmıştır. Oluşturulan derin öğrenme modellerinin yapıları, eğitim için kullanılan parametre değerleri, kullanılan katmanlar, doğrulama verileri için elde edilen karmaşıklık matrisleri, doğruluk ve kayıp grafikleri ayrıntılı olarak gösterilmiştir. Çalışmamız 3 adet konvolüsyon katmanı, 3 adet batch normalizasyon katmanı, 2 adet maxpooling katmanı, 1 flatten, 2 droupout, 2 dense katmanından oluşan farklı bir ağ yapısı ile gerçekleştirilmiştir. Ayrıca önerdiğimiz modelle görüntü sınıflandırılmasının farklı veriler üzerindeki performansı artırılmıştır. Test sonunda çalışmamız çeşitli değerlendirme metriklerine göre doğruluk sonuçları karşılaştırılmıştır. Kullanılan tüm veri setleri için en iyi ve en kötü bulunan görüntüler tespit edilmiştir. Önerilen CNN modeli ile MNIST ve Fashion MNIST veri setleri için yüksek doğruluk oranları gözlemlenmiş olup bu değerler sırasıyla 99.22 ve 99.21dir.

Makale detayları

A new Convolutional Neural Network proposal for classifying various images

Yazarlar:

Yeşim Tiraki | Hasan Temurtaş | Soydan Serttaş | Çiğdem Bakır

Yayıncı:

Nigde Omer Halisdemir University

Ana Başlıklar:

Derin Öğrenme

Dergi:

Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.13 Issue 1, pp.262-278

Belge Tipi:

Makale

Yayın Tarihi:

1/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001309732

pISSN:

2564-6605

Deep learning

Convolutional Neural Networks

CIFAR-10

CIFAR-100

MNIST

Fashion MNISTDerin öğrenme

Evrişimli Sinir Ağı

CIFAR-10

CIFAR-100

MNIST

Fashion MNIST

Sayı

Makale detayları

Yazarlar: Yeşim Tiraki | Hasan Temurtaş | Soydan Serttaş | Çiğdem Bakır

Yayıncı: Nigde Omer Halisdemir University

Ana Başlıklar: Derin Öğrenme

Dergi: Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.13 Issue 1, pp.262-278

Belge Tipi: Makale

Yayın Tarihi: 1/2024

Belge Dili: Türkçe

Document ID (Org No.): 2001309732

pISSN: 2564-6605

Sayı

Yazarlar:

Yeşim Tiraki | Hasan Temurtaş | Soydan Serttaş | Çiğdem Bakır

Yayıncı:

Nigde Omer Halisdemir University

Ana Başlıklar:

Derin Öğrenme

Dergi:

Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi Vol.13 Issue 1, pp.262-278

Belge Tipi:

Makale

Yayın Tarihi:

1/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001309732

pISSN:

2564-6605