Sürdürülebilir Kalkınma Amaçları
Yazarlar
- Yildirim, Ozal(19)
- Utku, Anıl(11)
- Talo, Muhammed(10)
- Utku, Anil(10)
- Baloglu, Ulas Baran(9)
- Acharya, U. Rajendra(7)
- Ergen, Burhan(7)
- Doğan, Gürkan(5)
- Karakose, Mehmet(5)
- Yilmazer, Merve(5)
- Can, Umit(4)
- Demir, Yakup(4)
- Imak, Andaç(4)
- Acharya, U Rajendra(3)
- Aydin, Galip(3)
- Ertargin, Merve(3)
- Gunay, Mihriban(3)
- Tan, Ru-San(3)
- Yıldırım, Özal(3)
- Akcayol, M. Ali(2)
- Ay, Betul(2)
- Barut, Abdulkadir(2)
- Iliev, Teodor(2)
- Karabatak, Murat(2)
- Kaya, Sema Kayapinar(2)
- Korkmaz, Adem(2)
- Kosunalp, Selahattin(2)
- Kutlu, Banu(2)
- Orhan, Ahmet(2)
- Pilatin, Abdulmuttalip(2)
- Rajendra Acharya, U.(2)
- Sengur, Abdulkadir(2)
- Tan, Ru San(2)
- Türkoğlu, Muammer(2)
- Çelebi, Adalet(2)
- Çelik, Yusuf(2)
- Şengür, Abdulkadir(2)
- Agdas, Mehmet T.(1)
- Agdas, Mehmet Tevfik(1)
- Akcayol, Muhammet Ali(1)
Yayınlar
- MakaleAçık Erişim2025Scopus SJRQ1Hybrid GA-ConvLSTM for data-driven prediction of climate variables: a case study of the most biodiverse cities in India
Utku, Anıl; Akyol, Sinem
Climate variables affect ecosystems’ structure, sustainability, and biodiversity. Monitoring and analysis of climate variables are essential for the management of ecosystems and the conservation of biodiversity. Temperature affects the distributions, habitats, and phonological events of living things in ecosystems. Humidity is essential for meeting plants’ water needs and the habitat’s quality and health. Dew point is essential for water quality, balance, plant health, and microclimate. Wind speed is essential for pollinating plants, microhabitats, and the health of living things. For such reasons, accurate prediction of climate variables using artificial intelligence methods can effectively predict the future of ecosystems and biodiversity and develop future strategic plans. This study aimed to predict the climate variables of temperature, humidity, dew point, and wind speed of Bhubaneshwar, Dehradun, Guwahati, Mysore, and Nagpur, among the cities with the highest biodiversity in India. It aimed to determine the long-term impacts of climate change on managing ecosystems and biodiversity protection. The hyperparameters of the ConvLSTM model, developed using CNN and LSTM, were optimized with a genetic algorithm, and the GA-ConvLSTM hybrid model was created. GA-ConvLSTM was tested with XGBoost, SVM, ANFIS, CNN, LSTM, and base ConvLSTM using approximately 15 years of hourly, up-to-date, and real-time climate variable data of cities. This work is one of the pioneering studies using GA to optimize deep learning hybrid models for multi-city climate prediction through extensive long-term high-resolution datasets. The experimental results reveal that GA-ConvLSTM outperforms conventional models, producing an R2 score exceeding 0.9 for each city across all climate variables. GA-ConvLSTM produced temperature predictions with 0.833 RMSE and 0.995 R2 for Bhubaneswar, humidity predictions with 3.461 RMSE and 0.990 R2, dew point predictions with 0.632 RMSE and 0.995 R2, and wind speed predictions with 2.278 RMSE and 0.993 R2. The experimental findings confirm the improved performance over conventional models due to SVM and XGBoost failing to achieve these R2 values and high RMSE scores. The research meets three principal obstacles because of the combination of stochastic climate datasets and complex nonlinear patterns, which also encounter limitations in conventional methods for managing scale-dependency relationships and manual parameter adjustments. This approach delivers three main strengths because it combines CNN and LSTM networks through genetic programming while tracking temporal and spatial relationships and processing noisy data automatically, leading to improved results compared to other methods. Unlike other published models in the field, the hybrid model GA-ConvLSTM combines optimized CNN and LSTM elements with genetic algorithm elements. The proposed model received its first application in this research for multi-city climate prediction through training with 15 years of hourly data. Standard methods encounter challenges when detecting long-term dependencies, but GA-ConvLSTM successfully addresses this issue.
AtıfScopus1OpenAlex1
İncele - MakaleAçık Erişim2025Scopus SJRQ2HASK-Net: new hybrid attention selective kernel network for automatic colon cancer detection from colonoscopy images
Imak, Andaç; Doğan, Gürkan; Sengur, Abdulkadir; Ergen, Burhan
Colorectal cancer is a frequently analyzed cancer type seen today. It is the third leading of cancer in terms of incidence. It is among the second leading cases resulting in death. However, when detecting pre-cancerous polyps using traditional methods, the rate of being overlooked is high. Therefore, the rate of missed polyps can be greatly reduced with deep learning-based computer-aided diagnosis (CADx) methods. In this study, a novel convolutional neural network based on CADx, called HASK-Net, is proposed for automatic polyp detection. While HASK-Net uses ResNet50 as a feature extractor in the encoder network, the novely designed hybrid attention selective kernel convolution uses it in the decoder network to increase the representation power in feature maps and learn more complex features. In experimental results on the publicly available Hiper-Kvasir and EndoTech 2020 benchmark datasets, HASK-Net showed extremely promising performance with a dice similarity score of 97.02% and 84.83% and an mIoU score of 95.38% and 84.32%, respectively.
- MakaleAçık Erişim2025Scopus SJRQ1Deep learning and optimization-based feature selection for fetal health classification using CTG data
Kaya, Turgay; Kaya, Duygu; Atar, Fatmanur
This study introduces a DL and metaheuristic optimization-based framework for fetal health assessment using cardiotocography (CTG) signals to mitigate maternal and neonatal mortality. One-dimensional CTG signals were transformed into 2D representations, and deep feature extraction was performed using AlexNet. Feature vectors FC6, FC7, and their combination were subjected to optimization via Whale Optimization Algorithm (WOA + DL) and War Strategy Optimization (WSO + DL), utilizing updated fitness functions tailored for feature selection. Experimental results with SVM classifiers demonstrated superior performance with FC6 (89.98 %) and WSO + DL (90.17 %). FC6 exhibited strong discriminative capacity, while FC7 contained semantically richer features. The concatenated FC6 + FC7 vector increased feature diversity. WSO + DL achieved optimal balance across classification accuracy, feature subset size, convergence rate, and overall performance metrics. The integration of DL and metaheuristic algorithms effectively isolated informative feature subsets, improving training efficiency, minimizing redundant/noisy data, reducing overfitting risk, and enhancing classification accuracy. Optimization method selection proved critical to overall model performance.
AtıfScopus6WoS1OpenAlex6
İncele - MakaleAçık Erişim2025Scopus SJRQ2Species classification of microalgae using a CNN-based deep learning approach under optimal cultivation conditions
Simsek, Gokce Kendirlioglu; Ertargin, Merve; Pezzolesi, Laura; Pistocchi, Rossella; Yildirim, Ozal; Cetin, A. Kadri
Microalgae possess significant potential in a wide range of applications due to their valuable bioactive compounds. They are utilized in biofuel production to reduce dependence on fossil fuels, in wastewater treatment for the biological removal of heavy metals and pollutants, in carbon capture for mitigating climate change, and in the pharmaceutical and nutraceutical industries as supplements. As their applications expand, the accurate and efficient identification of microalgal species becomes increasingly important. Traditional classification methods are time-consuming and rely heavily on expert knowledge. The main aim of this study is to develop a reliable, fast, and expert-independent deep learning-based approach for the classification of microalgae species using microscopic images. In this context, deep learning techniques, specifically convolutional neural network (CNN)-based models were employed to classify microalgal species. Four widely used pre-trained CNN architectures (ResNet152, DenseNet201, MobileNetV2, and EfficientNetB0), along with a custom-designed CNN, were implemented. The models were trained and tested on a labeled dataset consisting of microscopic images of Chlorella vulgaris, Scenedesmus acutus, and Haematococcus pluvialis. The classification models achieved accuracy rates ranging from 96.87 % (Custom CNN) to 100 % (DenseNet201), demonstrating the potential of CNN-based approaches in automating and improving microalgae species identification.
AtıfScopus5WoS5OpenAlex5İncele - MakaleAçık Erişim2025Scopus SJRQ2Advancing Air Quality Monitoring: Deep Learning-Based CNN-RNN Hybrid Model for PM2.5 Forecasting
Utku, Anıl; Can, Umit; Alpsülün, Mustafa; Balıkçı, Hasan Celal; Amoozegar, Azadeh; Pilatin, Abdulmuttalip; Barut, Abdulkadir
Particulate matter, particularly PM2.5, poses a significant threat to public health due to its ability to disperse widely and its detrimental impact on the respiratory and circulatory systems upon inhalation. Consequently, it is imperative to maintain regular monitoring and assessment of particulate matter levels to anticipate air pollution events and promptly mitigate their adverse effects. However, predicting air quality is inherently complex, given the multitude of variables that influence it. Deep learning models, renowned for their ability to capture nonlinear relationships, offer a promising approach to address this challenge, with hybrid architectures demonstrating enhanced performance. This study aims to develop and evaluate a hybrid model integrating Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) for forecasting PM2.5 levels in India, Milan, and Frankfurt. A comparative analysis with established deep learning and machine learning techniques substantiates the superior predictive capabilities of the proposed CNN-RNN model. The findings underscore its potential as an effective tool for air quality prediction, with implications for informed decision-making and proactive intervention strategies to safeguard public health.
AtıfScopus10WoS6OpenAlex11İncele - MakaleAçık Erişim2025Scopus SJRQ2A New Hybrid ConvViT Model for Dangerous Farm Insect Detection
Utku, Anil; Kaya, Mahmut; Canbay, Yavuz
This study proposes a novel hybrid convolution and vision transformer model (ConvViT) designed to detect harmful insect species that adversely affect agricultural production and play a critical role in global food security. By utilizing a dataset comprising images of 15 distinct insect species, the suggested approach combines the strengths of traditional convolutional neural networks (CNNs) with vision transformer (ViT) architectures. This integration aims to capture local-level morphological features effectively while analyzing global spatial relationships more comprehensively. While the CNN structure excels at discerning fine morphological details of insects, the ViT’s self-attention mechanism enables a holistic evaluation of their overall configurations. Several data preprocessing steps were implemented to enhance the model’s performance, including data augmentation techniques and strategies to ensure class balance. In addition, hyperparameter optimization contributed to more stable and robust model training. Experimental results indicate that the ConvViT model outperforms commonly used benchmark architectures such as EfficientNetB0, DenseNet201, ResNet-50, VGG-16, and standalone ViT, achieving a classification accuracy of 93.61%. This hybrid approach improves accuracy and strengthens generalization capabilities, delivering steady performance during training and testing phases, thereby increasing its reliability for field applications. The findings highlight that the ConvViT model achieves high efficiency in pest detection by integrating local and global feature learning. Consequently, this scalable artificial intelligence solution can support sustainable agricultural practices by enabling the early and accurate identification of pests and reducing the need for intensive pesticide use.
AtıfScopus9WoS5OpenAlex11İncele - MakaleYalnız künye2024Scopus SJRQ2Deep learning-assisted Hubble parameter analysis
Salti, Mehmet; Kangal, Evrim Ersin; Zengin, Bilgin
We turn our attention on evaluating the most recent Hubble parameter data measured via the differential evolution of cosmic-chronometers from a deep learning perspective. To achieve this goal, we start our investigation by introducing the selected theoretical setup and compiling the most recent statistical data obtained in cosmology experiments. Then we implement a tuned version of the long-short term memory (LSTM) architecture and run it to predict possible values of the Cosmic Hubble parameter for different red-shift states. Since we observe a good correlation between the observed and predicted datasets of the Hubble parameter, we conclude that the machine learning approaches can play important roles in the future cosmology investigations.
AtıfScopus2WoS2OpenAlex2İncele - MakaleYayıncıda açık erişim2024Scopus SJRQ2Mechanical and electrical faults detection in induction motor across multiple sensors with CNN-LSTM deep learning model
Ertargin, Merve; Yildirim, Ozal; Orhan, Ahmet
The utilization of monitoring sensors in machinery has led to the mainstream adoption of fault detection and diagnosis in time series data across various industrial applications. Deep learning techniques, specifically in constructing fault diagnosis models by extracting insights from historical equipment fault data, are receiving widespread attention as crucial tools in ensuring the safety and reliability of motor systems. In this study, a CNN-LSTM-based deep learning model is proposed for the detection of electric motor faults. Three distinct sets of accelerometer sensor data are provided as input to the model, enabling a comprehensive evaluation of its performance across various sensor configurations. The model demonstrated a remarkable capacity for generalization, achieving impressive accuracy rates of 99.96% for Accelerometer-1, 98.88% for Accelerometer-2, and 99.37% for Accelerometer-3. This underscores the robustness and adaptability of the proposed CNN-LSTM model in effectively detecting electric motor faults regardless of the specific accelerometer sensor employed.
AtıfScopus47OpenAlex38İncele - MakaleAçık Erişim2024Scopus SJRQ1A new hybrid mobile CNN approach for crosswalk recognition in autonomous vehicles
Doğan, Gürkan; Ergen, Burhan
While automobile transportation is increasing worldwide, it also negatively affects the safety of road users. Along with the neglect of traffic rules, pedestrians account for 22% of all highway traffic deaths. Millions of pedestrians suffer non-fatal injuries from these accidents. Most of these injuries and deaths occur at crosswalks, where the highway and pedestrians intersect. In this study, deep learning-based a new hybrid mobile CNN approaches are proposed to reduce injuries and deaths by automatically recognizing of crosswalks in autonomous vehicles. The first of these proposed approaches is the HMCNet approach, which is a hybrid model in which the MobileNetv3 and MNasNet CNN models are used together. This model achieves approximately 2% more accuracy than the peak performance of the lean used MobileNetv3 and MNasNet models. Another proposed approach is the FHMCNet approach, which increases the success of the HMCNet approach. In the FHMCNet approach, LSVC feature selection method and SVM classification method are used in addition to HMCNet. This approach increased the classification success of HMCNet by more than approximately 2%. Finally, the proposed FHMCNet offered approximately 3% more classification accuracy than state-of-the-art methods in the literature.
AtıfScopus10WoS7OpenAlex8İncele - MakaleAçık Erişim2024Scopus SJRQ1Spread patterns of COVID-19 in European countries: hybrid deep learning model for prediction and transmission analysis
Utku, Anıl; Akcayol, M. Ali
The COVID-19 pandemic has profoundly impacted healthcare systems and economies worldwide, leading to the implementation of travel restrictions and social measures. Efforts such as vaccination campaigns, testing, and surveillance have played a crucial role in containing the spread of the virus and safeguarding public health. There needs to be more research exploring the transmission dynamics of COVID-19, particularly within European nations. Therefore, the primary objective of this research was to examine the spread patterns of COVID-19 across various European countries. Doing so makes it possible to implement preventive measures, allocate resources, and optimize treatment strategies based on projected case and mortality rates. For this purpose, a hybrid prediction model combining CNN and LSTM models was developed. The performance of this hybrid model was compared against several other models, including CNN, k-NN, LR, LSTM, MLP, RF, SVM, and XGBoost. The empirical findings revealed that the CNN-LSTM hybrid model exhibited superior performance compared to alternative models in effectively predicting the transmission of COVID-19 within European nations. Furthermore, examining the peak of case and death dates provided insights into the dynamics of COVID-19 transmission among European countries. Chord diagrams were drawn to analyze the inter-country transmission patterns of COVID-19 over 5-day and 14-day intervals.
AtıfScopus6WoS3OpenAlex7İncele - BildiriAçık Erişim2024A Deep Learning-Based Approach for Damage Detection in Cultural Heritage Images and Generating Heat Maps
Ogdu, Cagatay Umut; Yilmazer, Merve; Karakose, Mehmet
Cultural structures may be damaged and damaged over time due to external factors such as natural and social disasters, which may lead to the loss of national assets if cultural assets are not protected. Therefore, the protection of cultural heritage is of great importance not only for today's generations but also for future. In recent years, deep learning and image processing have made great progress and have become very effective for detecting and analyzing patterns. These techniques can also be used to solve complex problems such as damage detection of structures in the field of cultural heritage. Deep learning methods can be used to determine the current status of structures and detect possible damage by working on images obtained from different angles of structures, while also reducing human errors. In this study, as an important step for the protection of cultural heritage, a detection method was developed using deep learning models on damaged and intact cultural heritage building images. The data set used in the study consists of damaged and intact cultural heritage building images. An attempt was made to determine the model that gave the best results by using pre-trained models with the Transfer Learning method. Heat maps were created using the Grad-CAM method to identify damaged areas on the trained models, thus making it possible to more clearly identify the damaged areas in the image. This method can play an important role in preserving cultural heritage. It can be used to detect damaged areas and direct them to repair.
AtıfScopus3OpenAlex4İncele - MakaleAçık Erişim2024Scopus SJRQ1A new hybrid approach for grapevine leaves recognition based on ESRGAN data augmentation and GASVM feature selection
Doğan, Gürkan; Imak, Andaç; Ergen, Burhan; Sengur, Abdulkadir
Grapevine leaf is a commodity that is collected only once a year and has a high return on investment due to its export. However, only certain types of grapevine leaves are consumed. Therefore, it is extremely important to distinguish the types of grapevine leaves. In particular, performing this process automatically on industrial machines will reduce human errors, workload, and thus cost. In this study, a new hybrid approach based on a convolutional neural network is proposed that can automatically distinguish the types of grapevine leaves. In the proposed approach, firstly, the overfitting of network models is prevented by applying data augmentation techniques. Second, new synthetic images were created with the ESRGAN technique to obtain detailed texture information. Third, the top blocks of the MobileNetV2 and VGG19 CNN models were replaced with the newly designed top block, effectively extracting features with the data. Fourthly, the GASVM algorithm was adapted and used to create a subset of the features to eliminate the ineffective and unimportant ones from the obtained features. Finally, SVM classification was performed with the feature subset consisting of 314 features, and approximately 2% higher accuracy and MCC score were obtained compared to the approaches in the literature.
AtıfScopus8WoS6OpenAlex9İncele - BildiriAçık Erişim2024A New Approach to Detecting Free Parking Spaces Based on YOLOv6 and Keyframe Detection with Video Analytics
Yilmazer, Merve; Karakose, Mehmet; Tanberk, Senem; Arslan, Samed
With urbanization and population growth, automobile use has also increased. As a result of this, there is a need to expand parking management and inspection systems to ensure safe and efficient use of parking areas. A new deep learning-based method has been proposed as an alternative to traditional systems to control free spaces in parking areas monitored by CCTV security cameras. A large amount of video data needs to be processed for parking areas recorded with video cameras 24 hours a day, 365 days a year. Repeating frames in videos cause overlearning in deep learning models. It also increases memory and computation cost. For this reason, keyframe detection based on sampling was first performed on the sample video of the Dragon Lake Parking (DLP) dataset and a summary of the dataset was obtained. Then, empty parking areas were detected by YOLOv6 deep neural network. It has been observed that the model can detect empty parking areas at a rate of 96.6% mAP.
AtıfScopus7OpenAlex5İncele - MakaleAçık Erişim2024Scopus SJRQ1A new CNN-based semantic object segmentation for autonomous vehicles in urban traffic scenes
Doğan, Gürkan; Ergen, Burhan
Semantic segmentation is the most important stage of making sense of the visual traffic scene for autonomous driving. In recent years, convolutional neural networks (CNN)-based methods for semantic segmentation of urban traffic scenes are among the trending studies. However, the methods developed in the studies carried out so far are insufficient in terms of accuracy performance criteria. In this study, a new CNN-based semantic segmentation method with higher accuracy performance is proposed. A new module, the Attentional Atrous Feature Pooling (AAFP) Module, has been developed for the proposed method. This module is located between the encoder and decoder in the general network structure and aims to obtain multi-scale information and add attentional features to large and small objects. As a result of experimental tests with the CamVid data set, an accuracy value of approximately 2% higher was achieved with a mIoU value of 70.59% compared to other state-of-art methods. Therefore, the proposed method can semantically segment objects in the urban traffic scene better than other methods.
AtıfScopus19WoS10İncele - MakaleYayıncıda açık erişim2024Scopus SJRQ3Fastener and rail surface defects detection with deep learning techniques
Yilmazer, Merve; Karakose, Mehmet
The railways, which are frequently used by countries for both passenger and freight transportation, should be checked periodically. Controls made with classical methods are slow and there are often overlooked faults. This work suggests a novel deep learning-based technique for identifying fastener and railway track surface defects. Within the scope of the proposed method, firstly, the railroad track was observed using an autonomous drone. Shaky images in the recorded video were removed with a video stabilization algorithm. Frames were created and labeled from the video, and rail and fastener regions were detected using the Faster R-CNN deep neural network. Fault detection was performed through ResNet101v2-based classification using different datasets for identifying surface detects in rails and different datasets for the detection of fasteners. The proposed method was experimentally shown to have a 98% accuracy rate for detecting rail surface flaws and a 95% accuracy rate for detecting fastener flaws. A user interface was developed to display the identified faulty images on computers, tablets, and mobile phones via a mobile application. The system, which was previously proposed in a different study, was retrained by going through the video stabilization step, thus improving the fault detection rate, and the method was also included in the user interface module. This study contributes to the processing of ever-increasing video data with deep learning-based methods. It is also anticipated that it will benefit researchers working in the field of railway non-contact fault detection.
AtıfScopus6OpenAlex6İncele - MakaleAçık Erişim2024Scopus SJRQ2Hybrid Deep Learning Model for Earthquake Time Prediction
Utku, Anıl; Akcayol, Muhammet Ali
Earthquakes are one of the most dangerous natural disasters that have constantly threatened humanity in the last decade. Therefore, it is extremely important to take preventive measures against earthquakes. Time estimation in these dangerous events is becoming more specific, especially in order to minimize the damage caused by earthquakes. In this study, a hybrid deep learning model is proposed to predict the time of the next earthquake to potentially occur. The developed CNN+GRU model was compared with RF, ARIMA, CNN and GRU. These models were tested using an earthquake dataset. Experimental results show that the CNN+GRU model performs better than others according to MSE, RMSE, MAE and MAPE metrics. This study highlights the importance of predicting earthquakes, providing a way to help take more effective precautions against earthquakes and potentially minimize loss of life and material damage. This study should be considered an important step in the methods used to predict future earthquakes and supports efforts to reduce earthquake risks.
AtıfScopus6WoS5TR Dizin4
İncele - MakaleAçık Erişim2024Scopus SJRQ2Detection of Threats to Farm Animals Using Deep Learning Models: A Comparative Study
Korkmaz, Adem; Agdas, Mehmet Tevfik; Kosunalp, Selahattin; Iliev, Teodor; Stoyanov, Ivaylo
The increasing global population and environmental changes pose significant challenges to food security and sustainable agricultural practices. To overcome these challenges, protecting farm animals and effectively detecting potential environmental threats is critical for economic and ecological sustainability. In this context, the current study examined the animal detection capabilities and efficiency of advanced deep learning models, such as YOLOv8, Yolo-NAS, and Fast-RNN, across a dataset of 2462 images encompassing various animal species that could pose a risk to farm animals. After converting the images into a standardized format, they were divided into three sets for training, validation, and testing, and each model was evaluated on this dataset during the analysis process. The findings indicated that the YOLOv8 model demonstrated superior performance, with 93% precision, 85.2% recall, and 93.1% mAP50 values, while Yolo-NAS was particularly noteworthy for its high recall value, indicating a remarkable detection ability. The Fast-RNN model also offered significant efficiency with balanced performance. The results reveal the considerable potential of deep learning-based object detection technologies in protecting farm animals and enhancing farm security. Additionally, this study provides valuable insights for future model optimization and customization research.
AtıfScopus9WoS7OpenAlex7İncele - MakaleAçık Erişim2023Deep Learning Based Video Event Classification
Gencaslan, Serim; Utku, Anil; Akcayol, M. Ali
In recent years, due to the growth of digital libraries and video databases, automatic detection of activities from videos and obtaining patterns from large datasets have come to the fore. Object detection from images is used as a tool for various applications and is the basis of video classification. Objects in videos are more difficult to identify than in single images, as the information in videos has a time-continuity constraint. Following the developments in the field of computer vision, the use of open source software packages for machine learning and deep learning and the developments in hardware technologies have enabled the development of new approaches. In this study, a deep learning-based classification model has been developed for the classification of sports branches in video. In the model developed using CNN, transfer learning has been applied with VGG-19. Experimental studies on 32827 frames using CNN and VGG-19 models showed that VGG-19 has a more successful classification performance than CNN with an accuracy rate of 83%.
AtıfWoS1OpenAlex3TR Dizin1İncele - BildiriAçık Erişim2023Yolov5 Based Fault Detection Approach in Railway Components
Yilmazer, Merve; Karakose, Mehmet
For safe transportation on the railway, the faults occurring in its components must be detected and repaired. Recently, fault detection methods have been developed on railway visual data using methods such as image processing, machine learning and deep learning. Especially since deep learning-based algorithms have a self-learning structure, the data used in model training and testing significantly affect the performance of the model. In the proposed method within the scope of this study, railway visual data were obtained using autonomous drone. YOLOv5 deep neural network is trained to detect faults that may occur in various components such as rail, fastener, sleeper and ballast in the railway. In order to increase the efficiency of the model, images obtained under different lighting conditions were included in the data set. In addition, by using the OpenCV library, the video stabilization process and the vibrations that occur during data acquisition in the video are eliminated. Thus, with the proposed method, it has been shown that the errors due to many components in the railway can be detected with a mAP rate of 94.5%.
AtıfScopus5OpenAlex3İncele - MakaleAçık Erişim2023A new approach based on convolutional neural network and feature selection for recognizing vehicle types
Doğan, Gürkan; Ergen, Burhan
The number of vehicles used in traffic life has reached enormous dimensions today. The increase in the number of vehicles day by day causes some traffic problems along with it; such as traffic congestion, accidents, pollution, and safety. To overcome all these problems, convolutional neural networks (CNN) methods are one of the trend methods used in recent years due to their success. In this study, a new approach is proposed to use this power of CNN in low-power devices. First of all, MobileNetv1, MobileNetv2, and NASNetMobile models were optimized to increase accuracy performance. Then, an approach is proposed in which these optimized mobile CNN approaches are used only as feature extractors, and methods such as combining, selecting, and classifying the obtained features are used together. As a result of the classification made with this approach, the classification accuracy has increased by approximately 5%.
AtıfScopus10OpenAlex12İncele
