Hüseyin Aydilek

Dr. Hüseyin Aydilek

Electrical and Electronics Engineering, Kırıkkale University, Kırıkkale, Turkey (Türkiye)
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 December 2022 in Uluslararası Muhendislik Arastirma ve Gelistirme Dergisi
Verified
Journal Article
Closed access
MEMustafa Yasin Erten
Mustafa Yasin Erten
Hüseyin AydilekHüseyin Aydilek
Hüseyin Aydilek
Solar power prediction is an important problem that has gained significant attention in recent years due to the increasing demand for renewable energy sources. In this paper, we present the results of using four different regression models for solar power prediction: linear regression, logistic regression, Lasso regression, and elastic regression. Our results show that all four models are able to accurately predict solar power, but Lasso regression and elastic regression outperform linear and logistic regression in terms of predicting the maximum solar power output. We also discuss the advantages and disadvantages of each model in the context of solar power prediction.

19 November 2024 in Engineering Applications of Artificial Intelligence
Verified
Journal Article
Closed access
MBMerve Bulut
Merve Bulut
Hüseyin AydilekHüseyin Aydilek
Hüseyin Aydilek
MEMustafa Yasin Erten
Mustafa Yasin Erten
Evrencan ÖzcanEvrencan Özcan
Evrencan Özcan

30 June 2025 in International Journal of Pure and Applied Sciences
Verified
Journal Article
Open access
PDF file
Hüseyin AydilekHüseyin Aydilek
Hüseyin Aydilek
MEMustafa Yasin Erten
Mustafa Yasin Erten
This study examines the effectiveness of deep learning-based models in the classification and monitoring of fish species. A dataset obtained from the Kaggle platform, containing 31 different fish species, was used to train MobileNetV2, DenseNet121, and VGG19 models. Bayesian optimization was applied to enhance model performance and determine the optimal hyperparameters. The results indicate that models trained with Bayesian optimization achieved significantly higher accuracy compared to those trained with randomly assigned hyperparameters. Additionally, the ensemble learning approach, which combined the outputs of individual models, yielded the best classification performance. This study demonstrates that deep learning techniques serve as a crucial tool for marine ecosystem conservation and sustainable fisheries practices.

31 December 2021 in International Scientific and Vocational Studies Journal
Verified
Journal Article
Open access
PDF file
Hüseyin AydilekHüseyin Aydilek
Hüseyin Aydilek
NINihat İnanç
Nihat İnanç
Due to its spectral properties, hyperspectral imaging is superior to other types of imaging tools in identifying, distinguishing and classifying objects. Hyperspectral imaging instruments can detect light reflected from certain wavelengths between infrared and ultraviolet, apart from the wavelength that the human eye can distinguish on the electromagnetic spectrum. While this feature provides detailed information about the spectral feature of the object under investigation, it causes its spatial resolution to be low due to the technical overlap between spatial resolution and spectral resolution. Today, applications of hyperspectral images are increasing in important fields such as agriculture, mining, medicine and pharmacy, especially for military purposes. In order for applications to produce more precise results, high spatial resolution is required, as well as high spectral information. Hardware solving of low spatial resolution problem is a difficult and costly method. Therefore, software solution is an interesting area in the field of image processing. In this thesis, a hybrid solution method based on deep learning and sparse representation is proposed to increase the low spatial resolution of hyperspectral images. The method obtains a super-resolution image from a single hyperspectral image with a low spatial image with a deep convolutional mesh. Later, the super-resolution image obtained and the original low-spatial-resolution hyperspectral image are fused with the dictionary learning method, resulting in a new super-resolution image with high spectral and spatial resolutions. The application results show that our method achieves successful results compared to many super resolution applications in the literature.

1 January 2023 in Uluslararası Muhendislik Arastirma ve Gelistirme Dergisi
Verified
Journal Article
Closed access
PDF file
FYFikret Yalçinkaya
Fikret Yalçinkaya
ASAta Sevinç
Ata Sevinç
Hüseyin AydilekHüseyin Aydilek
Hüseyin Aydilek
AEAli Erbaş
Ali Erbaş
In this paper, an ultrasonic therapy device using fuzzy-logic(UTD-FL) has been designed, constructed and tested with phantom materials. Fuzzy-logic rules have been determined using four parameters. In clinical practice, ultrasonic therapy is conducted solely based on subjective evaluation of medical experts, but in UTD-FL, fuzzy-logic rules decide automatically the three critically important characteristic parameters of applicable output power to the patient, namely the power of the signal, the percent of the duty-cycle and the signal frequency. The signal frequency and specifically its amplitude have critical effect on the temperature rise of the tissue test-point or surface. Therefore the intensity of the ultrasonic frequency and the duration of treatment-time are absolutely vital, that is why this instrument has been developed. This instrument is expected to prevent possible side effects, injuries, and potential damages on real tissues due to experts’ uneasiness. The test results of this newly developed medical device have been compared with clinical-practice. The instrument produces optimum output power due to its fuzzy-logic rules based mode design, and IR-temperature sensor based feedback effect; whereas the clinical mode inputs only clinical experience base gained medical data.