Yazar "Sert, Eser" seçeneğine göre listele
Listeleniyor 1 - 14 / 14
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe A hybrid deep learning approach combining neutrosophic set theory and multi-axis vision transformer for nutrient deficiency classification in plants(Pergamon-Elsevier Science Ltd, 2025) Sert, Eser; Kiziloluk, SonerPlant nutrient deficiencies in agricultural production, especially nitrogen and potassium deficiencies, threaten global food security. The fact that traditional detection methods are time-consuming, costly, and require expertise increases the need for automatic and reliable classification systems. Current deep learning models lack the ability to assess the reliability of their predictions, which can lead to misclassifications in low-quality or ambiguous images. In order to overcome these limitations, an innovative approach that combines Neutrosophic Set theory with the Multi-Axis Vision Transformer (MaxViT), one of the latest Vision Transformer (ViT) models, namely Neutrosophic-MaxViT, is proposed in this study. The proposed model increases the explicability and reliability of classification results by integrating MaxViT's hybrid architecture, which includes local convolutions and global attention mechanisms, with the truth (T), indeterminacy (I), and falsity (F) components of Neutrosophic Set theory. In addition, a special Neutrosophic Entropy Loss Function has been developed that optimizes the T, I, and F components. The performance of the model was evaluated on the Early Nutritional Stress Detection of Plants (EarlyNSD) dataset, which includes healthy, nitrogen, and potassium-deficient conditions of ashgourd, bittergourd, and snakegourd plants. The experimental results show that the Neutrosophic-MaxViT model outperforms advanced models, including MaxViT, Regular Networks Y (RegNetY), Convolutional Neural Network Next (ConvNeXt), and ViT-Tiny, achieving 95.31 % accuracy and a macro F1-score of 0.9529. This study contributes to artificial intelligence-based agricultural solutions, enhancing productivity and offering a reliable framework for sustainable practices.Öğe A new hybrid approach based on AOA, CNN and feature fusion that can automatically diagnose Parkinson's disease from sound signals: PDD-AOA-CNN(Springer London Ltd, 2024) Yildirim, Muhammed; Kiziloluk, Soner; Aslan, Serpil; Sert, EserParkinson's is one of the most rapidly increasing neurological diseases in the world, caused by the deficiency of dopamine-producing cells in the brain. Voice disorders are a significant finding in the early stage of Parkinson's disease (PD). Detection of this finding at an early stage of the disease allows early treatment of the disease. Therefore, in this study, using sound data, a hybrid model for detecting PD has been designed. In the developed method, first of all, the sound data were converted into spectrograms. Then, the feature maps of the obtained spectrogram images were extracted using 3 different CNN architectures. Feature maps with different features obtained by utilizing the accumulation of different architectures were combined. Then, these features were selected using the arithmetic optimization algorithm (AOA), one of the most recent metaheuristic optimization algorithms, and then classified by support vector machine (SVM) and K-nearest neighbors (KNN). One of the important novelties in the study is the reduction of the size of the acquired feature maps with AOA, a new and high-performance metaheuristic approach. The success of the proposed model in diagnosing Parkinson's disease reached up to 98.19%. In addition, feature maps of the sound data in the dataset were acquired by using the MFCC method to compare the performance of the proposed model. Eight different classifiers were used to categorize the acquired feature maps. The highest accuracy value obtained in this method was obtained in the Random Forest classifier with 93.98%.Öğe A new super resolution Faster R-CNN model based detection and classification of urine sediments(Elsevier, 2023) Avci, Derya; Sert, Eser; Dogantekin, Esin; Yildirim, Ozal; Tadeusiewicz, Ryszard; Plawiak, PawelThe diagnosis of urinary tract infections and kidney diseases using urine microscopy images has gained significant attention of medical community in recent years. These images are usually created by physicians' own rule of thumb manually. However, this man-ual urine sediment analysis is usually labor-intensive and time-consuming. In addition, even when physicians carefully examine an image, an erroneous cell recognition may occur due to some optical illusions. In order to achieve cell recognition in low-resolution urine microscopy images with a higher level of accuracy, a new super resolution Faster Region-based Convolutional Neural Network (Faster R-CNN) method is proposed. It aims to increase resolution in low-resolution urine microscopy images using self-similarity based single image super resolution which was used during the pre-processing. De-noising based Wiener filter and Discrete Wavelet Transform (DWT) are used to de-noise high resolution images, respectively, to increase the level of accuracy for image recognition. Finally, for the feature extraction and classification stages, AlexNet, VGFG16 and VGG19 based Faster R-CNN models are used for the recognition and detection of multi-class cells. The model yielded accuracy rates are 98.6%, 96.4% and 96.2% respectively.(c) 2022 Nalecz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences. Published by Elsevier B.V. All rights reserved.Öğe Automatic detection of harmful cyanobacterial genera using deep CNN models and artemisinin optimization(Nature Portfolio, 2025) Topaloglu, Fatih; Kiziloluk, Soner; Sert, Eser; Yildirim, MuhammedConcerns over the spread of Cyanobacteria, which can lead to dangerous blooms that harm drinking water quality and, therefore, the health of plants and animals, are being raised by global warming. Traditional methods for assessing the amount of toxic species in water samples are often time-consuming, require intensive manual effort, are prone to subjective errors, and can lead to delays in necessary water management interventions. This emphasizes the pressing need for a quick and precise automated method. Both aquatic and terrestrial environments include cyanobacteria, and under some circumstances, poisonous cyanobacteria can grow in large numbers and form harmful blooms called harmful cyanobacterial blooms (Cyano-HABs). In addition, cyanoHABs cause hypoxia, ecological imbalances, the generation of toxins, and other detrimental phenomena that put people, animals, and plants in danger of illness. Climate change is expected to cause these situations to increase in frequency and globally. This study presents a novel approach for the automatic detection of harmful cyanobacteria genera by utilizing a newly introduced and publicly available dataset, TCB-DS. In the initial stage, discriminative features are extracted using two powerful deep Convolutional Neural Network (CNN) models: ShuffleNet and ResNet-50. Subsequently, feature fusion is applied to the extracted features to enhance the representation. Then, to select the most relevant features, feature selection is performed using the Artemisinin Optimization (AO) algorithm, a robust meta-heuristic algorithm inspired by the mechanisms of malaria treatment and recently proposed in 2024. This step aims to reduce feature redundancy and improve the overall efficiency of the model. In classifying microscopic images of cyanobacteria species with the proposed method, GoogleNet, MobileNetV2, EfficientNetb0, DarkNet53, ShuffleNet, and ResNet101 models were used. Among these, the proposed method obtained the highest accuracy, with a mean accuracy of 97.471% and max accuracy of 97.683%. Since these results are the highest accuracy values obtained in the TCB-DS dataset, our proposed method significantly improves water quality monitoring in our world.Öğe Comparative Analysis of Machine Learning, Deep Learning, and Transformer-Based Models for Fake News Detection(Institute of Electrical and Electronics Engineers Inc., 2025) Gürer, Aybuke; Sert, EserThe widespread of fake news on digital and social media has alarming consequences for public perception, political stability, and social confidence. This study provides a comparative evalution of various machine learning, deep learning, and transformer-based models. The more than 25,000 labeled news articles in the publicly accessible ISOT Fake and Real News Dataset served as the basis for all tests. Eight different models, such as Support Vector Machine (SVM), Convolutional Neural Network (CNN), and BERT, were created by splitting the data into 80% training and 20% test groups. Feature extraction was performed and trained using TF-IDF feature vectors for traditional models, text sequential representations for deep learning models, and contextual embeddings for transformer models. Eight different models, such as Support Vector Machine (SVM), Convolutional Neural Network (CNN), and BERT, were created by splitting the data into 80% training and 20% test groups. The CNN model (99.84% accuracy and 1.0000 ROC-AUC) achieved the highest accuracy and the best ROC-AUC scores. The CNN model is followed by BERT (99.65% accuracy), which has a complex architecture. However, simpler models such as Logistic Regression and SVM also yielded surprising results with accuracy exceeding 99%. These results suggest the importance of data quality and feature representation for fake news detection. © 2025 IEEE.Öğe An effective Turkey marble classification system: Convolutional neural network with genetic algorithm -wavelet kernel - Extreme learning machine(IIETA International Information and Engineering Technology Association, 2021) Avcı, Derya; Sert, EserMarble is one of the most popular decorative elements. Marble quality varies depending on its vein patterns and color, which are the two most important factors affecting marble quality and class. The manual classification of marbles is likely to lead to various mistakes due to different optical illusions. However, computer vision minimizes these mistakes thanks to artificial intelligence and machine learning. The present study proposes the Convolutional Neural Network- (CNN-) with genetic algorithm- (GA) Wavelet Kernel- (WK-) Extreme Learning Machine (ELM) (CNN–GA-WK-ELM) approach. Using CNN architectures such as AlexNet, VGG-19, SqueezeNet, and ResNet-50, the proposed approach obtained 4 different feature vectors from 10 different marble images. Later, Genetic Algorithm (GA) was used to optimize adjustable parameters, i.e. k, 1, and m, and hidden layer neuron number in Wavelet Kernel (WK) – Extreme Learning Machine (ELM) and to increase the performance of ELM. Finally, 4 different feature vector parameters were optimized and classified using the WK-ELM classifier. The proposed CNN–GA-WK-ELM yielded an accuracy rate of 98.20%, 96.40%, 96.20%, and 95.60% using AlexNet, SequeezeNet, VGG-19, and ResNet-50, respectively.Öğe Ensemble Residual Network Features and Cubic-SVM Based Tomato Leaves Disease Classification System(International Information and Engineering Technology Association, 2022) Özyurt, Fatih; Sert, Eser; Avcı, DeryaThe need for automatic disease detection applications that can help farmers in the detection of agricultural product diseases is increasing day by day. Convolutional Neural Network (CNN) is a very popular field in image processing, recognition, and classification. It is seen that CNN architectures are used in the determination of agricultural products. In this study, 3 different ResNet architectures of the features automatically are used in the detection of tomato diseases. The most efficient features obtained from these architectures have been obtained by the NCA algorithm again. The features obtained have been trained with the Cubic SVM machine learning algorithm. Tomato leaves belonging to a total of 10 classes have been trained at 80% and a test performance rate of 98.2% has been achieved.Öğe EO-CNN: Equilibrium Optimization-Based hyperparameter tuning for enhanced pneumonia and COVID-19 detection using AlexNet and DarkNet19(Elsevier, 2024) Kiziloluk, Soner; Sert, Eser; Hammad, Mohamed; Tadeusiewicz, Ryszard; Plawiak, PawelConvolutional neural networks (CNN) have been increasingly popular in image categorization in recent years. Hyperparameter optimization is a critical stage in enhancing the effectiveness of CNNs and achieving better results. Properly tuning hyperparameters allows the model to exhibit improved performance and facilitates faster learning. Misconfigured hyperparameters can prolong the training time or lead to the model not learning at all. Manually tuning hyperparameters is a time-consuming and challenging process. Automatically adjusting hyperparameters helps save time and resources. This study aims to propose an approach that shows higher classification performance than unoptimized convolutional neural network models, even at low epoch values, by automatically optimizing the hyperparameters of AlexNet and DarkNet19 with equilibrium optimization, the newest metaheuristic algorithm. In this respect, the proposed approach optimizes the number and size of filters in the first five convolutional layers in AlexNet and DarkNet19 using an equilibrium optimization algorithm. To evaluate the efficacy of the suggested method, experimental analyses were conducted on the pneumonia and COVID-19 datasets. An important advantage of this approach is its ability to accurately classify medical images. The testing process suggests that utilizing the proposed approach to optimize hyperparameters for AlexNet and DarkNet19 led to a 7% and 4.07% improvement, respectively, in image classification accuracy compared to nonoptimized versions of the same networks. Furthermore, the approach displayed superior classification performance even in a few epochs compared to AlexNet, ShuffleNet, DarkNet19, GoogleNet, MobileNet-V2, VGG-16, VGG-19, ResNet18, and Inceptionv3. As a result, automatic tuning of the hyperparameters of AlexNet and DarkNet-19 with EO enabled the performance of these two models to increase significantly.Öğe Hurricane-Faster R-CNN-JS: Hurricane detection with faster R-CNN using artificial Jellyfish Search (JS) optimizer(SPRINGER, 2022) Kızıloluk, Soner; Sert, EserA hurricane is a type of storm called tropical cyclone (TC) and is likely to lead to severe storms and heavy rains. An early detection of hurricanes using satellite images can alarm people about upcoming disasters and thus minimize any casualties and material losses. Faster R-CNN is one of the most popular and recent object detection approaches. In the present study, AlexNet hyperparameters, which is a CNN model used as a feature extractor in Faster R-CNN, were optimized using artificial Jellyfish Search (JS), which is a recent algorithm, in order to propose a Faster R-CNN with a higher performance. The proposed approach is called Hurricane-Faster R-CNN-JS, since it is used as an early hurricane detection approach on satellite images before these hurricanes reach the land. The results of the present study demonstrated that hyperparameter optimization increased the detection performance of the proposed approach by 10% compared to AlexNet without optimized hyperparameters. As feature extractors of Faster R-CNN, the present study benefited from various architectures such as MobileNet-V2, GoogLeNet, AlexNet, ResNet 18, ResNet 50, VGG-16 and VGG-19 without any optimized hyperparameters to compare them with the proposed approach. It was observed that Average Precision (AP) of Hurricane-Faster R-CNN-JS was 97.39%, which was a remarkably higher AP level compared to other approaches.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 MidFusionEfficientV2: Improving Ophthalmic Diagnosis with Mid-Level RGB-LBP Fusion and SE Attention(Mdpi, 2026) Keles, Julide Kurt; Kiziloluk, Soner; Sert, Eser; Talo, Furkan; Yildirim, MuhammedBackground/Objectives: Early diagnosis of eye diseases is critically important for enhancing individuals' quality of life and reducing the risk of vision loss. In this study, a deep learning-based hybrid model called MidFusionEfficientV2 has been proposed to classify eye diseases, including uveitis, conjunctivitis, cataract, eyelid drooping, and normal conditions. Methods: The model presents a dual-branch architecture that combines an RGB image branch with an EfficientNetV2-S architecture and a specialized texture branch based on Local Binary Pattern (LBP) transformation at an intermediate level. Thanks to the Squeeze-and-Excitation (SE) blocks integrated into the LBP branch, channel-based attention mechanisms have been activated, enhancing the prominence of textural features. The features obtained from the RGB and LBP branches were combined at an intermediate level and transferred to the classification stage. Results: Experimental studies on the five-class eye disease dataset from the Mendeley Data platform have shown that the proposed model outperformed six strong models (ResNetV2, ConvNeXt, DenseNet-121, EfficientNet-B1, MobileNetV3 Large, and EfficientNetV2-S) with an accuracy of 98%. Especially in the difficult-to-diagnose uveitis class, recall and F1 scores of 97% and 94%, respectively, were achieved. Conclusions: The results show that a moderate combination of color and texture features significantly improves classification performance, and that MidFusionEfficientV2 offers a reliable and effective solution for the automatic diagnosis of eye diseases.Öğe A new 3D segmentation approach using extreme learning machine algorithm and morphological operations(Elsevier, 2020) Kaya, Ertuğrul; Sert, EserSegmentation is one of the most crucial steps of image processing. Because 3D images contain depth information, they have gradually gained importance for numerical systems in image analysis. In the present study, a new 3D segmentation method based on extreme learning machine and morphological operations (3DS-ELM) is proposed. The present study benefits from extreme learning machine (ELM) algorithm, which is a novel and fast learning algorithm for single-hidden layer feedforward networks (SLFNs), for training objects. Because a 3D model contains many points, direct segmentation on a 3D model is time-consuming and causes problems in the segmentation process, the proposed approach minimizes these problems and offers a quick and high-performance 3D segmentation method that can be used in various industrial fields. The proposed 3DS-ELM was compared with different approaches in order to analyze its 3D segmentation performance. Experimental studies proved that the proposed 3DS-ELM performed better than other approaches.Öğe TSA-CNN-AOA: Twitter sentiment analysis using CNN optimized via arithmetic optimization algorithm(Springer London Ltd, 2023) Aslan, Serpil; Kiziloluk, Soner; Sert, EserCOVID-19, a novel virus from the coronavirus family, broke out in Wuhan city of China and spread all over the world, killing more than 5.5 million people. The speed of spreading is still critical as an infectious disease, and it causes more and more deaths each passing day. COVID-19 pandemic has resulted in many different psychological effects on people's mental states, such as anxiety, fear, and similar complex feelings. Millions of people worldwide have shared their opinions on COVID-19 on several social media websites, particularly on Twitter. Therefore, it is likely to minimize the negative psychological impact of the disease on society by obtaining individuals' views on COVID-19 from social media platforms, making deductions from their statements, and identifying negative statements about the disease. In this respect, Twitter sentiment analysis (TSA), a recently popular research topic, is used to perform data analysis on social media platforms such as Twitter and reach certain conclusions. The present study, too, proposes TSA using convolutional neural network optimized via arithmetic optimization algorithm (TSA-CNN-AOA) approach. Firstly, using a designed API, 173,638 tweets about COVID-19 were extracted from Twitter between July 25, 2020, and August 30, 2020 to create a database. Later, significant information was extracted from this database using FastText Skip-gram. The proposed approach benefits from a designed convolutional neural network (CNN) model as a feature extractor. Thanks to arithmetic optimization algorithm (AOA), a feature selection process was also applied to the features obtained from CNN. Later, K-nearest neighbors (KNN), support vector machine, and decision tree were used to classify tweets as positive, negative, and neutral. In order to measure the TSA performance of the proposed method, it was compared with different approaches. The results demonstrated that TSA-CNN-AOA (KNN) achieved the highest tweet classification performance with an accuracy rate of 95.098. It is evident from the experimental studies that the proposed approach displayed a much higher TSA performance compared to other similar approaches in the existing literature.Öğe UC-merced image classification with CNN feature reduction using wavelet entropy optimized with genetic algorithm(International Information and Engineering Technology Association, 2020) Özyurt, Fatih; Ava, Engin; Sert, EserThe classification of high-resolution and remote sensed terrain images with high accuracy is one of the greatest challenges in machine learning. In the present study, a novel CNN feature reduction using Wavelet Entropy Optimized with Genetic Algorithm (GA-WEE-CNN) method was used for remote sensing images classification. The optimal wavelet family and optimal value of the parameters of the Wavelet Sure Entropy (WSE), Wavelet Nom Entropy (WNE), and Wavelet Threshold Entropy (WTE) were calculated, and given to classifiers such as K-Nearest Neighbors (KNN) and Support Vector Machine (SVM). The efficiency of the proposed hybrid method was tested using the UC-Merced dataset. 80% of the data were used as training data, and a performance rate of 98.8% was achieved with SVM classifier, which has been the highest ratio compared to all studies using same dataset so far with only 18 features. These results proved the advantage of the proposed method.
