A5080395937- Automated Road and Building Extraction
- Neural Networks and Applications
- Remote Sensing and LiDAR Applications
- Face recognition and analysis
- Remote-Sensing Image Classification
- Advanced Vision and Imaging
- Distributed systems and fault tolerance
- COVID-19 diagnosis using AI
- Advanced Image and Video Retrieval Techniques
- EEG and Brain-Computer Interfaces
- Face and Expression Recognition
- Advanced Image Processing Techniques
- Biometric Identification and Security
- Blind Source Separation Techniques
- Advanced Data Storage Technologies
- Domain Adaptation and Few-Shot Learning
- Video Surveillance and Tracking Methods
- Fault Detection and Control Systems
- Distributed and Parallel Computing Systems
- Online and Blended Learning
- 3D Shape Modeling and Analysis
- Advanced Database Systems and Queries
- Image Retrieval and Classification Techniques
- Robotics and Sensor-Based Localization
- Digital Media Forensic Detection
Middle East Technical University
2012-2023
Bitlis Eren University
2018
Ankara University
2014
Scientific and Technological Research Council of Turkey
1996-2002
<h3>Importance</h3> Application of deep learning algorithms to whole-slide pathology images can potentially improve diagnostic accuracy and efficiency. <h3>Objective</h3> Assess the performance automated at detecting metastases in hematoxylin eosin–stained tissue sections lymph nodes women with breast cancer compare it pathologists' diagnoses a setting. <h3>Design, Setting, Participants</h3> Researcher challenge competition (CAMELYON16) develop solutions for node (November 2015-November...
Signal classification is an important issue in brain computer interface (BCI) systems. Deep learning approaches have been used successfully many recent studies to learn features and classify different types of data. However, the number that employ these on BCI applications very limited. In this study we aim use deep methods improve performance EEG motor imagery signals.In investigate convolutional neural networks (CNN) stacked autoencoders (SAE) Motor Imagery signals. A new form input...
The automatic detection of airports is essential due to the strategic importance these targets. In this letter, a runway method based on textural properties proposed since they are most descriptive element an airport. Since best discriminative features for airport runways cannot be trivially predicted, Adaboost algorithm employed as feature selector over large set features. Moreover, selected with corresponding weights can provide information hidden characteristics runways. Thus,...
In this paper, a neural network structure based on self organizing feature maps (SOFM) is proposed for fingerprint classification. order to be able deal with images having distorted regions, the SOFM learning and classification algorithms are modified. For purpose, concept of "certainty" introduced used in modified algorithms. This classifier together identifier, constitute subsystems an automated identification system, named HALafis. Our results show that trained sufficiently large...
In this letter, a two-stage method for airport detection on remote sensing images is proposed. the first stage, new algorithm composed of several line-based processing steps used extraction candidate regions. second scale-invariant feature transformation and Fisher vector coding are efficient representation nonairport regions support machines employed classification. order to evaluate performance proposed method, extensive experiments conducted airports around world with different layouts....
While the early diagnosis of hematopoietic system disorders is very important in hematology, it a highly complex and time consuming task. The requires lot patients to be followed-up by experts which, general unfeasible because required number experts. differential blood counter (DBC) that we have developed an attempt automate task performed manually routine. In our system, cells are segmented using active contour models (snakes balloons), which initialized morphological operators. Shape...
The differential blood counter system we developed is an attempt to automate the task performed manually by experts in routine. Feature extraction and classification are two important components of our automated system. In this paper, cells using various approaches including neural network based classifiers support vector machine presented together with features used classification.
In this paper, we present a novel automatic approach based on local shape descriptors to discriminate 3-D facial scans of different individuals. Our begins with registration, smoothing and uniform resampling face data. Then, uniformly resampled data are used generate index, curvedness, gaussian mean curvature values each point the Hence obtain 2-D matrices representing geometry information. SIFT applied high dimensional feature vector having information is obtained. Finally, projected low...
Human level recall performance in detecting breast cancer considering microcalcifications from mammograms has a value between 74.5% and 92.3%. In this research, we approach to microcalcification classification problem using convolutional neural networks along with various preprocessing methods such as contrast scaling, dilation, cropping etc. decision fusion ensemble of networks. Various experiments on Digital Database for Screening Mammography dataset showed that poses great importance the...
Sleep spindles are 2 hallmark of the stage sleep. Their distribution over non-REM sleep is clinically important. In this paper, a method that detects in EEG proposed. Short time Fourier transform used for feature extraction. Both multilayer perceptron and Support Vector Machine utilized detection comparison. The classification performance MLP found to be 88.7% SVM as 95.4%. It should noted there might differences also visual scoring by experts, so results obtained quite satisfactory.
Deep models trained on large-scale RGB image datasets have shown tremendous success. It is important to apply such deep real-world problems. However, these suffer from a performance bottleneck under illumination changes. Thermal IR cameras are more robust against changes, and thus can be very useful for the In order investigate efficacy of combining feature-rich visible spectrum thermal modalities, we propose an unsupervised domain adaptation method which does not require RGB-to-thermal...
An inferential control methodology, that utilizes an artificial neural network (ANN) estimator for a model predictive controller (MPC), is developed industrial multicomponent distillation column. In the of product compositions by feedback system, because difficulty on-line measurements compositions, temperature can be utilized. The selection measurement points done help singular value decomposition (SVD) analysis together with column dynamics information. A moving window ANN designed to...
This letter introduces a new method for building extraction in satellite images. The algorithm first identifies the shadow segments on an oversegmented image, and then neighboring segments, which are assumed to be cast by single building, merged. Next, candidate regions where buildings most likely occur detected using these regions. Along with this information, closeness shadows illumination direction spectral properties of used classify them as belonging "building" or not. Then,...
METU INteroperable Database System (MIND) is a multidatabase system that aims at achieving interoperability among heterogeneous, federated DBMSs. MIND architecture if based on OMG distributed object management model. It implemented top of CORBA compliant ORB, namely, ObjectBroker. provides users single ODMG-93 common data model, and global query language SQL. This makes it possible to incorporate both relational oriented databases into the system. Currently Oracle 7, Sybase OODBMS (MOOD)...
Brain Computer Interface (BCI) systems provide control of external devices by using only brain activity. In recent years, there has been a great interest in developing BCI for different applications. These are capable solving daily life problems both healthy and disabled people. One the most important applications is to communication people that totally paralysed. this paper, parts system methods used each part reviewed. Neuroimaging devices, with an emphasis on EEG (electroencephalography),...
This study investigated the use of deep convolutional neural networks (CNNs) in providing a solution for problem airport detection remote sensing images (RSIs). In recent years, Deep CNNs have gained much attention with numerous applications having been undertaken area computer vision. Researchers generally approach as pattern recognition problem, which first various distinctive features are extracted, and then classifier is adopted to detect airports. not only ensure tuned feature vector,...
Advances in hardware and pattern recognition techniques, along with the widespread utilization of remote sensing satellites, have urged development automatic target detection systems satellite images. Automatic airports is particularly essential, due to strategic importance these targets. In this paper, a runway method using segmentation process based on textural properties proposed for airport runways, which most distinguishing element an airport. Several local features are extracted...
A method called optimistic with dummy locks (ODL) is suggested for concurrency control in distributed databases. It shown that by using long-term locks, the need information about write sets of validated transactions eliminated and, during validation test, only related sites are checked. The to be aborted immediately recognized before reducing costs restarts. Usual read and used as short-term test. use approach eliminates system-wide critical section results a parallel performance ODL...
In this study, the neuron of random neural network (RNN) model (Gelenbe 1989) is designed using digital circuitry. RNN model, each accumulates arriving pulses and can fire if its potential at a given instant time strictly positive. Firing occurs random, intervals between successive firing instants following an exponential distribution constant rate. When fires, it routes generated to appropriate output lines in accordance with connection probabilities. circuitry fundamental parts are...