Ulus Çevik
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ulus Çevik has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Gökay Dişken
Adana Alparslan Türkeş Science and Technology University
3 shared publications - Zekeriya Tufekci
Çukurova University
3 shared publications - Abidin Çalışkan
Batman Üniversitesi
2 shared publications Erdal Dayak
Department of Informatics, University of Gaziantep, Gaziantep, Turkey
2 shared publications- Turgay Ibrikci
Çukurova University
2 shared publications - Ömer Cengiz Çelebi
Department of Electrical and Electronics Engineering, University of Çukurova, Adana, Turkey
2 shared publications
Latest Publications
- 17 October 2022
- Published by Cukurova Universitesi Muhendislik-Mimarlik Fakultesi Dergisi in Çukurova Üniversitesi Mühendislik Fakültesi Dergisi
- Vol. 37 (3) , 699-706
A Design of Fully Automated Irrigation System IoT-Based Approach for Greenhouses
Turkey is a country with a high potential for agriculture purposes. Irrigation is a need to receive agricultural output. Crop yield increases with optimal irrigation. Therefore, this work has designed a fully automated irrigation system with an IoT-based approach for greenhouses. Using 10HS soil moisture and pt1000 temperature sensors, the system has successfully generated irrigation-based decisions and saved 25 tons/da water compared to Et (Evapotranspiration) controller-based irrigation while plants received optimal irrigation. The system works online and has two operation modes: fully automated and manual. Fully automated system decreased water-wastages and labouring costs. Switching between modes can be operated using a mobile application and parameters of the system communicates via cloud services.
- 17 October 2022
- Published by Cukurova Universitesi Muhendislik-Mimarlik Fakultesi Dergisi in Çukurova Üniversitesi Mühendislik Fakültesi Dergisi
- Vol. 37 (3) , 699-706
A Design of Fully Automated Irrigation System IoT-Based Approach for Greenhouses
Turkey is a country with a high potential for agriculture purposes. Irrigation is a need to receive agricultural output. Crop yield increases with optimal irrigation. Therefore, this work has designed a fully automated irrigation system with an IoT-based approach for greenhouses. Using 10HS soil moisture and pt1000 temperature sensors, the system has successfully generated irrigation-based decisions and saved 25 tons/da water compared to Et (Evapotranspiration) controller-based irrigation while plants received optimal irrigation. The system works online and has two operation modes: fully automated and manual. Fully automated system decreased water-wastages and labouring costs. Switching between modes can be operated using a mobile application and parameters of the system communicates via cloud services.
- 28 April 2022
- Published by Wiley in Expert Systems
- Vol. 39 (8)
Analysis of breast cancer classification robustness with radiomics feature extraction and deep learning techniques
Breast cancer and breast imaging diagnostic procedures are typically carried out using a variety of imaging modalities, including mammography, MRI, and ultrasound. However, ultrasound and mammography have limitations and MRI is recognized as better than other procedures. Recent computational approaches, such as radiomics, applied to image analysis have shown remarkable progress in lowering diagnostic difficulties. This research analysed the robustness of breast tumour classification with feature extraction (radiomics) and a featureless method (deep learning). The proposal consists of two stages: the first stage introduced and explored radiomics‐based steps. A total of 111 tumour lesions were used to derive 74 radiomic features consisting of shape, and three separate second‐order metrics. Associations of these features were used to classify tumour lesions with four different kernels from support vector machine algorithm. In the confusion matrix analysis, the SVM‐RBF kernel developed optimal diagnostic efficiency with a maximum test accuracy of 97.06% on the combination of feature analysis. The second stage developed with deep learning techniques (InceptionV3 and CNN‐SVM). A total of 2998 images were used to create the models. In this portion, the CNN‐SVM model achieved the highest accuracy, 95.28%, with an AUC of 0.974, where the pre‐trained InceptionV3 achieved an AUC of only 0.932. Finally, the obtained result in both stages was discussed together and other related studies.
- 28 April 2022
- Published by Wiley in Expert Systems
- Vol. 39 (8)
Analysis of breast cancer classification robustness with radiomics feature extraction and deep learning techniques
Breast cancer and breast imaging diagnostic procedures are typically carried out using a variety of imaging modalities, including mammography, MRI, and ultrasound. However, ultrasound and mammography have limitations and MRI is recognized as better than other procedures. Recent computational approaches, such as radiomics, applied to image analysis have shown remarkable progress in lowering diagnostic difficulties. This research analysed the robustness of breast tumour classification with feature extraction (radiomics) and a featureless method (deep learning). The proposal consists of two stages: the first stage introduced and explored radiomics‐based steps. A total of 111 tumour lesions were used to derive 74 radiomic features consisting of shape, and three separate second‐order metrics. Associations of these features were used to classify tumour lesions with four different kernels from support vector machine algorithm. In the confusion matrix analysis, the SVM‐RBF kernel developed optimal diagnostic efficiency with a maximum test accuracy of 97.06% on the combination of feature analysis. The second stage developed with deep learning techniques (InceptionV3 and CNN‐SVM). A total of 2998 images were used to create the models. In this portion, the CNN‐SVM model achieved the highest accuracy, 95.28%, with an AUC of 0.974, where the pre‐trained InceptionV3 achieved an AUC of only 0.932. Finally, the obtained result in both stages was discussed together and other related studies.
- 30 March 2022
- Published by Anadolu Universitesi Bilim ve Teknoloji Dergisi-A: Uygulamali Bilimler ve Muhendislik in Eskişehir Technical University Journal of Science and Technology A - Applied Sciences and Engineering
- Vol. 23 (1) , 87-108
COMPARATIVE OF SUCCESS OF KNN WITH NEW PROPOSED K-SPLIT METHOD AND STRATIFIED CROSS VALIDATION ON REMOTE HOMOLOGUE PROTEIN DETECTION
In this study, a remote homologous protein detection problem, which is a problem belonging to the field of bioinformatics, which has a great contribution in the field of medicine, is discussed. Protein sequences taken from the SCOP database, which is an important and widely used database for proteins, were tested for remote homolog protein detection in this study. Feature vectors were obtained from the protein sequences using the bag of word model. These obtained feature vectors were classified using the kNN classifier algorithm. In this classification, the different distances were used as Bray Curtis, Chebyshev, Cosine, Dice, Euclidean, Hamming, Jaccard, Kulczynski, Matching coefficient, Minkowski, RogersTanimoto, RussellRao and SokalMichener on kNN classifier for remote homolog protein detection. There is proposed special k fold value formula for prevent imbalanced data problem. It has observed that the kNN algorithm with the Bray Curtis distance with cross validation with special k fold value shows the most successful performance with 99% accuracy.
- 30 March 2022
- Published by Anadolu Universitesi Bilim ve Teknoloji Dergisi-A: Uygulamali Bilimler ve Muhendislik in Eskişehir Technical University Journal of Science and Technology A - Applied Sciences and Engineering
- Vol. 23 (1) , 87-108
COMPARATIVE OF SUCCESS OF KNN WITH NEW PROPOSED K-SPLIT METHOD AND STRATIFIED CROSS VALIDATION ON REMOTE HOMOLOGUE PROTEIN DETECTION
In this study, a remote homologous protein detection problem, which is a problem belonging to the field of bioinformatics, which has a great contribution in the field of medicine, is discussed. Protein sequences taken from the SCOP database, which is an important and widely used database for proteins, were tested for remote homolog protein detection in this study. Feature vectors were obtained from the protein sequences using the bag of word model. These obtained feature vectors were classified using the kNN classifier algorithm. In this classification, the different distances were used as Bray Curtis, Chebyshev, Cosine, Dice, Euclidean, Hamming, Jaccard, Kulczynski, Matching coefficient, Minkowski, RogersTanimoto, RussellRao and SokalMichener on kNN classifier for remote homolog protein detection. There is proposed special k fold value formula for prevent imbalanced data problem. It has observed that the kNN algorithm with the Bray Curtis distance with cross validation with special k fold value shows the most successful performance with 99% accuracy.
- 30 March 2021
- Published by The Scientific and Technological Research Council of Turkey (TUBITAK-ULAKBIM) - DIGITAL COMMONS JOURNALS in Turkish Journal of Electrical Engineering and Computer Sciences
- Vol. 29 (2) , 616-631
Ensemble learning of multiview CNN models for survival time prediction of brain tumor patients using multimodal MRI scans
- 30 March 2021
- Published by The Scientific and Technological Research Council of Turkey (TUBITAK-ULAKBIM) - DIGITAL COMMONS JOURNALS in Turkish Journal of Electrical Engineering and Computer Sciences
- Vol. 29 (2) , 616-631
Ensemble learning of multiview CNN models for survival time prediction of brain tumor patients using multimodal MRI scans
- Open Access
- 1 June 2018
- Published by University of Slavonski Brod in Tehnicki vjesnik - Technical Gazette
- Vol. 25 (3) , 679-686
An Efficient Noisy Pixels Detection Model for CT Images using Extreme Learning Machines
In this study, a new and rapid hidden resource decomposition method has been proposed to determine noisy pixels by adopting the extreme learning machines (ELM) method. The goal of this method is not only to determine noisy pixels, but...
- 1 June 2018
- Published by University of Slavonski Brod in Tehnicki vjesnik - Technical Gazette
- Vol. 25 (3) , 679-686
Open AccessAn Efficient Noisy Pixels Detection Model for CT Images using Extreme Learning Machines
In this study, a new and rapid hidden resource decomposition method has been proposed to determine noisy pixels by adopting the extreme learning machines (ELM) method. The goal of this method is not only to determine noisy pixels, but...
- Open Access
- 11 October 2017
- Published by Springer Nature in EURASIP Journal on Audio, Speech, and Music Processing
- Vol. 2017 (1) , 23
A robust polynomial regression-based voice activity detector for speaker verification
- 11 October 2017
- Published by Springer Nature in EURASIP Journal on Audio, Speech, and Music Processing
- Vol. 2017 (1) , 23
Open AccessA robust polynomial regression-based voice activity detector for speaker verification
- Open Access
- 1 March 2017
- Published by Walter de Gruyter GmbH in Archives of Acoustics
- Vol. 42 (1) , 127-135
Speaker Model Clustering to Construct Background Models for Speaker Verification
Conventional speaker recognition systems use the Universal Background Model (UBM) as an imposter for all speakers. In this paper, speaker models are clustered to obtain better imposter model representations for speaker verification purpose. First, a UBM is trained, and speaker models are adapted from the UBM. Then, the k-means algorithm with the Euclidean distance measure is applied to the speaker models. The speakers are divided into two, three, four, and five clusters. The resulting cluster centers are used as background models of their respective speakers. Experiments showed that the proposed method consistently produced lower Equal Error Rates (EER) than the conventional UBM approach for 3, 10, and 30 seconds long test utterances, and also for channel mismatch conditions. The proposed method is also compared with the i-vector approach. The three-cluster model achieved the best performance with a 12.4% relative EER reduction in average, compared to the i-vector method. Statistical significance of the results are also given.
- 1 March 2017
- Published by Walter de Gruyter GmbH in Archives of Acoustics
- Vol. 42 (1) , 127-135
Open AccessSpeaker Model Clustering to Construct Background Models for Speaker Verification
Conventional speaker recognition systems use the Universal Background Model (UBM) as an imposter for all speakers. In this paper, speaker models are clustered to obtain better imposter model representations for speaker verification purpose. First, a UBM is trained, and speaker models are adapted from the UBM. Then, the k-means algorithm with the Euclidean distance measure is applied to the speaker models. The speakers are divided into two, three, four, and five clusters. The resulting cluster centers are used as background models of their respective speakers. Experiments showed that the proposed method consistently produced lower Equal Error Rates (EER) than the conventional UBM approach for 3, 10, and 30 seconds long test utterances, and also for channel mismatch conditions. The proposed method is also compared with the i-vector approach. The three-cluster model achieved the best performance with a 12.4% relative EER reduction in average, compared to the i-vector method. Statistical significance of the results are also given.
- 1 January 2017
- Published by The Scientific and Technological Research Council of Turkey (TUBITAK-ULAKBIM) - DIGITAL COMMONS JOURNALS in Turkish Journal of Electrical Engineering and Computer Sciences
- Vol. 25 (1) , 200-208
Neville interpolation-based normal estimation
- 1 January 2017
- Published by The Scientific and Technological Research Council of Turkey (TUBITAK-ULAKBIM) - DIGITAL COMMONS JOURNALS in Turkish Journal of Electrical Engineering and Computer Sciences
- Vol. 25 (1) , 200-208
Neville interpolation-based normal estimation
- 8 June 2016
- Published by Taylor & Francis in IETE Technical Review
- Vol. 34 (3) , 321-332
A Review on Feature Extraction for Speaker Recognition under Degraded Conditions
Speech is a signal that includes speaker's emotion, characteristic specification, phoneme-information etc. Various methods have been proposed for speaker recognition by extracting specifications of a given utterance. Among them, short-term cepstral features are used excessively in speech, and speaker recognition areas because of their low complexity, and high performance in controlled environments. On the other hand, their performances decrease dramatically under degraded conditions such as channel mismatch, additive noise, emotional variability, etc. In this paper, a literature review on speaker-specific information extraction from speech is presented by considering the latest studies offering solutions to the aforementioned problem. The studies are categorized in three groups considering their robustness against channel mismatch, additive noise, and other degradations such as vocal effort, emotion mismatch, etc. For a more understandable representation, they are also classified into two tables by utilizing their classification methods, and used data-sets.
- 8 June 2016
- Published by Taylor & Francis in IETE Technical Review
- Vol. 34 (3) , 321-332
A Review on Feature Extraction for Speaker Recognition under Degraded Conditions
Speech is a signal that includes speaker's emotion, characteristic specification, phoneme-information etc. Various methods have been proposed for speaker recognition by extracting specifications of a given utterance. Among them, short-term cepstral features are used excessively in speech, and speaker recognition areas because of their low complexity, and high performance in controlled environments. On the other hand, their performances decrease dramatically under degraded conditions such as channel mismatch, additive noise, emotional variability, etc. In this paper, a literature review on speaker-specific information extraction from speech is presented by considering the latest studies offering solutions to the aforementioned problem. The studies are categorized in three groups considering their robustness against channel mismatch, additive noise, and other degradations such as vocal effort, emotion mismatch, etc. For a more understandable representation, they are also classified into two tables by utilizing their classification methods, and used data-sets.
- 14 December 2015
- Published by Tech Science Press in Intelligent Automation and Soft Computing
- Vol. 23 (1) , 95-102
Selective surface normal estimation for volume rendering
- 14 December 2015
- Published by Tech Science Press in Intelligent Automation and Soft Computing
- Vol. 23 (1) , 95-102
Selective surface normal estimation for volume rendering