Yazar "Aktas, Hakan" seçeneğine göre listele

Listeleniyor 1 - 5 / 5
  • Küçük Resim Yok
    Öğe
    Classification of hazelnut kernels with deep learning
    (Elsevier, 2023) Unal, Zeynep; Aktas, Hakan
    In this study a deep learning-based, highly accurate network structure was applied to discriminate the whole kernel of hazelnut from other classes such as damaged kernel, shell, and undersized, which are present in small quantities after mechanized sorting and later subject to manual sorting. An industrial setup was developed to generate datasets for these four classes, and 2094 images of each class were recorded. With the obtained datasets, EfficientNetB0- EfficientNetB1- EfficientNetB2-EfficientNetB3 and InceptionV3 network structures were first trained from scratch and the highest test accuracy was calculated as 97.85 % for the InceptionV3 network structure. To further increase in the test accuracy, the transfer learning structure was used by applying the Imagenet coefficients. As a result of the training using Imagenet weights, the highest test accuracy was calculated as 99.28 % for the EfficientNetB2 and EfficientNetB3 structures. Although both networks give the same result in the classification of four hazelnut classes, the success in discrimination of the whole kernel of hazelnut from other classes is a key solution to decrease economic loss due to incorrect classification. From this point of view, according to confusion matrixes, it was concluded that the EfficientNetB3 structure is four times more efficient than the EfficientNetB2 structure.
  • Küçük Resim Yok
    Öğe
    Classification of pistachios with deep learning and assessing the effect of various datasets on accuracy
    (Springer, 2022) Aktas, Hakan; Kizildeniz, Tefide; Unal, Zeynep
    Pistachio is a healthy and delicious snack with high economic value, especially when product consists of only open pistachios. For this reason, many studies have been carried out in the literature to classify Pistachio according to whether they are open or closed. In this study, the classification process of pistachios was carried out according to whether they are open or closed using deep learning techniques. The prominent aspect of the study is that the datasets obtained with an industrial experimental set-up are used in the training of the network in order to be suitable for industrial applications and to classify it with high accuracy. In this study, AlexNet and Inception V3 structure were trained and tested with this industrial data set, the test accuracy was calculated as 96.13% and 96.54%, respectively. In order to compare the industrial data set and the desktop data set, both data sets were created. As a result of training and testing the AlexNet structure with this desktop data set, the test accuracy was calculated as 100%. When the test images from industrial dataset are fed to the network structure trained with the desktop dataset, the test accuracy was obtained as 61.75%. On the contrary, when desktop data set is fed to Alexnet structure trained with industrial data set, test accuracy is calculated as 99.84%. This clearly demonstrates how accurately the industrial dataset performs in industrial classification applications, while the desktop dataset has poor accuracy in industrial applications.
  • Küçük Resim Yok
    Öğe
    Design Optimization for High-Performance Computing Using FPGA
    (Springer International Publishing Ag, 2024) Isik, Murat; Inadagbo, Kayode; Aktas, Hakan
    Reconfigurable architectures like Field Programmable Gate Arrays (FPGAs) have been used for accelerating computations in several domains because of their unique combination of flexibility, performance, and power efficiency. However, FPGAs have not been widely used for high-performance computing, primarily because of their programming complexity and difficulties in optimizing performance. We optimize Tensil AI's open-source inference accelerator for maximum performance using ResNet20 trained on CIFAR in this paper in order to gain insight into the use of FPGAs for high-performance computing. In this paper, we show how improving hardware design, using Xilinx Ultra RAM, and using advanced compiler strategies can lead to improved inference performance. We also demonstrate that running the CIFAR test data set shows very little accuracy drop when rounding down from the original 32bit floating point. The heterogeneous computing model in our platform allows us to achieve a frame rate of 293.58 frames per second (FPS) and a %90 accuracy on a ResNet20 trained using CIFAR. The experimental results show that the proposed accelerator achieves a throughput of 21.12 Giga-Operations Per Second (GOP/s) with a 5.21W on-chip power consumption at 100 MHz. The comparison results with off-the-shelf devices and recent state-of-the-art implementations illustrate that the proposed accelerator has obvious advantages in terms of energy efficiency.
  • Küçük Resim Yok
    Öğe
    Detection of bruises on red apples using deep learning models
    (Elsevier, 2024) Unal, Zeynep; Kizildeniz, Tefide; Ozden, Mustafa; Aktas, Hakan; Karagoz, Omer
    The detection and sorting of bruised apples after harvest play a crucial role in improving their economic value by eliminating surface defects. This also reduces the risk of contamination of infected apples during transport and storage. It can be done by using manual detection or machine vision techniques in red, green, and blue (RGB) colors to detect bruises on apples of various skin colors; however, in the early stages of bruising, it is challenging. Therefore, the main purpose of this study is determin of the effectiveness of Deep Learning models combined with the Near Infrared (NIR) imaging system for naturally bruised Super Chief red apples immediately after harvest. In total, 1000 images for the healthy class and 500 images for the bruised class were acquired from 500 apples. After the images were acquired with the RGB and NIR cameras, the data sets were divided into training (70 %), validation (15 %), and testing (15 %) sets. The Alexnet, the Inceptipon-V3, and the VGG16 network structures were trained using the training and validation data sets, and the trained network was evaluated using the test dataset. The VGG16 model achieved the highest test accuracy (86 %) when trained on the RGB data set, while the AlexNet model exhibited the lowest test accuracy (74.6 %). When the models were trained and tested with NIR datasets, 99.33 %, 100 % and 100 % accuracy rates were obtained for AlexNet, Inception V3, and VGG16, respectively. During the experiments, the VGG16 model trained with the NIR dataset achieved the lowest loss rate of 0.0002, whereas when trained and tested with the RGB dataset, the same VGG16 model also recorded the lowest loss rate of 0.353.These findings indicate that the deep learning models, particularly when trained with NIR data, demonstrate high accuracy rates in classifying apples as healthy or bruised, making them suitable for industrial classification applications. Therefore, the NIR data set is recommended for precise and reliable apple classification in industrial settings.
  • Küçük Resim Yok
    Öğe
    Harnessing the Power of Transfer Learning in Sunflower Disease Detection: A Comparative Study
    (Mdpi, 2023) Gulzar, Yonis; Unal, Zeynep; Aktas, Hakan; Mir, Mohammad Shuaib
    Sunflower is an important crop that is susceptible to various diseases, which can significantly impact crop yield and quality. Early and accurate detection of these diseases is crucial for implementing appropriate management strategies. In recent years, deep learning techniques have shown promising results in the field of disease classification using image data. This study presents a comparative analysis of different deep-learning models for the classification of sunflower diseases. five widely used deep learning models, namely AlexNet, VGG16, InceptionV3, MobileNetV3, and EfficientNet were trained and evaluated using a dataset of sunflower disease images. The performance of each model was measured in terms of precision, recall, F1-score, and accuracy. The experimental results demonstrated that all the deep learning models achieved high precision, recall, F1-score, and accuracy values for sunflower disease classification. Among the models, EfficientNetB3 exhibited the highest precision, recall, F1-score, and accuracy of 0.979. whereas the other models, ALexNet, VGG16, InceptionV3 and MobileNetV3 achieved 0.865, 0.965, 0.954 and 0.969 accuracy respectively. Based on the comparative analysis, it can be concluded that deep learning models are effective for the classification of sunflower diseases. The results highlight the potential of deep learning in early disease detection and classification, which can assist farmers and agronomists in implementing timely disease management strategies. Furthermore, the findings suggest that models like MobileNetV3 and EfficientNetB3 could be preferred choices due to their high performance and relatively fewer training epochs.