Yazar "Avci, Engin" seçeneğine göre listele
Listeleniyor 1 - 7 / 7
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe A hybrid ViT-CNN model with attention mechanism and dual feature selection for cervical cancer detection(Springer London Ltd, 2025) Onal, Merve Kesim; Van, Abdullah Enes; Avci, Derya; Yildirim, Muhammed; Bingol, Harun; Avci, EnginCervical cancer is the most common and fatal disease encountered by women worldwide. Early diagnosis of cervical cancer plays a critical role in reducing mortality rates and initiating the treatment process early. In this study, a hybrid model was developed to detect cervical cancer with high accuracy. In the developed model, 1280 and 768 features were extracted from each image, respectively, using pre-trained EfficientNet V2-M and Vision Transformer (ViT) architectures as the base; these features were combined to obtain a combined feature vector of 2048 dimensions. A feature attention mechanism was applied to highlight the important information in the data input, and then dimensionality reduction was performed using the mRMR and NCA methods. 838 standard features between the features selected with both methods were classified with six different machine learning algorithms. In the study, the five-class public SIPaKMeD dataset was used. The proposed model achieved a high accuracy value of 99.02% on the relevant dataset. In order to compare the performances of the models used in the study, different metrics such as Accuracy, Recall, Precision and F1 Score were evaluated. The proposed model was also compared with the performances of 8 different Convolutional Neural Network's (CNN's) and 4 different ViT architectures accepted in the literature. The proposed model produced more successful results than traditional approaches and similar studies in the literature.Öğe A Wavelet Enhanced Deep Learning Model for Mineral Image Classification: WFeedNet(Ieee-Inst Electrical Electronics Engineers Inc, 2026) Kesim Onal, Merve; Avci, EnginThe classification of minerals is of critical importance in many fields such as geology, mining, environmental engineering, and materials engineering. Accurate mineral identification directly impacts the efficiency of mineral exploration, ore enrichment, and industrial production processes. However, traditional identification methods performed in a laboratory setting (e.g., XRD, XRF, SEM-EDS, etc.) are costly, time-consuming, and dependent on expert knowledge. With the rapid advancements in artificial intelligence, deep learning-based image classification techniques, thanks to their ability to learn complex visual patterns automatically, have become powerful alternatives to traditional methods. In this study, a novel deep learning model called WFeedNet is proposed for the automatic classification of mineral images. The proposed model integrates both spatial and frequency domain information simultaneously by feeding the low-frequency components (LL) obtained from the multi-level wavelet transform (DWT) into the network via feed-forward blocks. Furthermore, the spatial and channel attention mechanisms integrated into the model ensure that feature maps are made more meaningful. Experimental results obtained from a unique dataset containing 1,474 mineral images demonstrate that the model achieved an accuracy of 94.8% without using any pretrained weights. The findings indicate that WFeedNet achieves higher classification success compared to traditional pretrained CNN and ViT-based models.Öğe Automated Early Detection of Skin Cancer Using a CNN-ViT-Attention-Based Hybrid Model(Mdpi, 2026) Kanat, Zekiye; Onal, Merve Kesim; Bingol, Harun; Sener, Serpil; Avci, Engin; Yildirim, MuhammedBackground/Objectives: Skin cancer is a very serious disease. There is a risk that the cancer will spread to other parts of the body as the cancerous tissue deepens. For this reason, early diagnosis is important because it allows for early initiation of treatment. This study proposes a hybrid model for the early diagnosis of skin cancer. Methods: The proposed model was developed using Convolutional Neural Networks (CNNs), Vision Transformer (ViT) architectures, and the k-Nearest Neighbors (KNN), Support Vector Machine (SVM), Naive Bayes (NB), Neural Network Classifiers, Decision Tree (DT), and Logistic Regression (LR) classifiers. Furthermore, the proposed model was fine-tuned to improve its disease diagnosis. Two attention mechanisms, channel and spatial, were used together in the proposed model. The HAM10000 dataset was used during the experiments. Class weighting was performed to ensure class-based balance in the dataset. Results: The proposed model was also compared with the CNN and ViT architectures frequently used in the literature. Among these models, the highest accuracy value of 95.1% was obtained with the proposed model. Conclusions: It is considered that the proposed model can be used as a decision support system for dermatologists in the diagnosis of skin cancer.Öğe Classification of Minerals Using Machine Learning Methods(Ieee, 2020) Onal, Merve Kesim; Avci, Engin; Ozyurt, Fatih; Orhan, AyhanMinerals are the structures formed by the combination of one or more elements as a result of geological process. The discipline which examines all aspects of minerals is called mineralogy. In mineralogy science, the definition and classification of minerals are made by taking into account many properties such as physical and chemical properties, internal crystal structures and optical properties. The classification of minerals has an important place in many engineering fields and such as geology, mining, geophysics, environment and in many fields such as mineral exploration, gemology, prospecting, ore processing. Both in the field environment and then in the laboratory environment, determination of all the properties of the mineral found in the field, is quite a long process. In addition, the researcher should have a good experience in the identification and classification process. In order to shorten this process in this study, it is aimed to classify minerals taken directly from the field and photographed, with AlexNet one of the convolutional neural network (CNN) architectures. In this study, 1491 images were used in 8 classes.Öğe Comparative Analysis of Conventional and Focused Data Augmentation Methods in Rib Fracture Detection in CT Images(Mdpi, 2025) Goktekin, Mehmet Cagri; Gul, Evrim; Aksu, Feyza; Gul, Yeliz; Ozen, Metehan; Salik, Yusuf; Avci, EnginBackground/Objectives: Rib fracture detection holds critical importance in the field of medical image processing. Methods: In this study, two different data augmentation methods, traditional data augmentation (Albumentations) and focused data augmentation (focused augmentation), were compared using computed tomography (CT) images for the detection of rib fractures on YOLOv8n, YOLOv8s, and YOLOv8m models. While the traditional data augmentation method applies general transformations to the entire image, the focused data augmentation method performs specific transformations by targeting only the fracture regions. Results: The model performance was evaluated using the Precision, Recall, mAP@50, and mAP@50-95 metrics. The findings revealed that the focused data augmentation method achieved superior performance in certain metrics. Specifically, analysis on the YOLOv8s model showed that the focused data augmentation method increased the mAP@50 value by 2.18%, reaching 0.9412, and improved the recall value for fracture detection by 5.70%, reaching 0.8766. On the other hand, the traditional data augmentation method achieved better results in overall precision metrics with the YOLOv8m model and provided a slight advantage in the mAP@50 value. Conclusions: The study indicates that focused data augmentation can contribute to achieving more reliable and accurate results in medical imaging applications.Öğe MFIF-DWT-CNN: Multi-focus image fusion based on discrete wavelet transform with deep convolutional neural network(Springer, 2024) Avci, Derya; Sert, Eser; Ozyurt, Fatih; Avci, EnginA new fusion method based on Multi-Focus Image Fusion Based on Discrete Wavelet Transform with Deep Convolutional Neural Network (MFIF-DWT-CNN) is presented to reduce spatial artifacts and blurring effects in edge details and increase the robustness of multifocal image fusion. The main purpose of the MFIF-DWT-CNN approach is to create a new merged image by collecting the required features from the main image. With the MFIF-DWT-CNN approach, information focused on individual images is combined into a single image, resulting in a clearer image. Within the scope of MFIF-DWT-CNN approach, DWT is applied to the image pairs and the obtained images are then given to the CNN architecture. The MFIF-DWT-CNN approach was developed in this study to reduce spatial artifacts and blurring effects in edge details and to increase the robustness of multifocal image fusion. In order to evaluate our proposed MFIF-DWT-CNN method, QMI, QG, QYi QCB evaluations were made on the public data set. From the experimental results, it is seen that the proposed method gives better results in the relevant metrics than the other methods. This demonstrated the effectiveness of the proposed method.Öğe Termal Kamera Görüntülerinin Çoklu Sınıflandırılması için Derin Öğrenme Tabanlı Tekniklerin Performans Karşılaştırılması(2025) ÖNAL, MERVE KESİM; USLU, Halil; Avci, Engin; Avci, DeryaTermal kameralar, cisimlerin sıcaklık farklılıklarını kızılötesi ışın değerlerine bağlı olarak renklendirdiği görüntüleme sistemleridir. Günümüzde başta savunma sanayi olmak üzere sağlık, ziraat, inşaat gibi birçok farklı alanda termal kameralar kullanılmaktadır. Özellikle savunma sanayi alanında kullanılan bu kameralardan elde edilen görüntüler, çeşitli nesnelerin ve canlıların tespiti için büyük önem arz etmektedir. Bu çalışmada termal kamera görüntülerinin sınıflandırması için derin öğrenme tabanlı tekniklerin kapsamlı bir karşılaştırması sunulmaktadır. Çalışmada 7 farklı Evrişimli Sinir Ağları mimarisi ile görüntülerin özellikleri çıkarılmış, 5 farklı sınıflandırma yöntemi ile sınıflandırılması sağlanmıştır. Performans değerlendirmesi için dengesiz çok sınıflı veri kümelerinin sınıflandırılmasına uygun metrikler olan dengeli doğruluk, makro ve mikro ortalama duyarlılık, makro ve mikro ortalama kesinlik, makro ve mikro ortalama F ölçütü metrik değerleri kullanılmıştır. Ayrıca tüm yapıların ayrı ayrı eğitim ve test süreleri karşılaştırılmıştır. Çalışmada en yüksek doğruluk değeri %95.24 ile Resnet101+Softmax ve Resnet50+DVM mimarilerinde elde edilmiştir. Sınıfların eşit ağırlıklı alındığı dengeli doğruluk değerinde ise en yüksek %95,17 ile Resnet101+Softmax mimarisinden elde edilmiştir. Resnet101+Softmax mimarisinde makro ortalamalı kesinlik 0.9579, makro ortalamalı F ölçütü 0.9543 ve mikro ortalamalı F ölçütü 0.9524 değerleri elde edilmiştir. Bu çalışma, küçük ve dengesiz termal görüntüler üzerinde, önceden eğitilmiş ESA ağlarının özellik çıkarımı ile makine öğrenimi sınıflandırıcılarının kullanımının, tamamen eğitilmiş ağlarla elde edilen performansa benzer sonuçlar sağlanabileceğini göstermiştir
