Recent studies have shown that lung cancer screening using annual low-dose computed tomography (C... more Recent studies have shown that lung cancer screening using annual low-dose computed tomography (CT) reduces lung cancer mortality by 20% compared to traditional chest radiography. Therefore, CT lung screening has started to be used widely all across the world. However, analyzing these images is a serious burden for radiologists. In this study, we propose a novel and simple framework that analyzes CT lung screenings using convolutional neural networks (CNNs) and reduces false positives. Our framework shows that even non-complex architectures are very powerful to classify 3D nodule data when compared to traditional methods. We also use different fusions in order to show their power and effect on the overall score. 3D CNNs are preferred over 2D CNNs because data are in 3D, and 2D convolutional operations may result in information loss. Mini-batch is used in order to overcome class-imbalance. Proposed framework has been validated according to the LUNA16 challenge evaluation and got scor...
Visual face tracking is one of the most important tasks in video surveillance systems. However, d... more Visual face tracking is one of the most important tasks in video surveillance systems. However, due to the variations in pose, scale, expression, and illumination it is considered to be a difficult task. Recent studies show that deep learning methods have a significant potential in object tracking tasks and adaptive feature selection methods can boost their performance. Motivated by these, we propose an end-to-end attentive deep learning based tracker, that is build on top of the state-of-the-art GOTURN tracker, for the task of real-time visual face tracking in video surveillance. Our method outperforms the state-of-the-art GOTURN and IVT trackers by very large margins and it achieves speeds that are very far beyond the requirements of real-time tracking. Additionally, to overcome the scarce data problem in visual face tracking, we also provide bounding box annotations for the G1 and G2 sets of ChokePoint dataset and make it suitable for further studies in face tracking under survei...
2016 20th National Biomedical Engineering Meeting (BIYOMUT), 2016
Laboratory mice are frequently used in biomedical studies. Facial expressions of mice provides im... more Laboratory mice are frequently used in biomedical studies. Facial expressions of mice provides important data about various issues. For this reason real time tracking of mice provide output to both researcher and software that operate on face image directly. Since body and face of laboratory mice is the same color and mice moves fast, tracking of face of mice is a challenging task. In recent years, methods that uses artificial neural networks provide effective solutions to problems such as classification, decision making and object recognition due to their ability to abstract training from data. In this study, a method based on deep learning is proposed for real time tracking of face of mice and successful results are obtained. Our studies are still going on in order to improve our results obtained using a limited dataset.
In this study, the effect of face regions on local shape descriptor based 3D face recognition was... more In this study, the effect of face regions on local shape descriptor based 3D face recognition was investigated. Our approach starts with calculation of the SIFT descriptors on the shape maps of the 3D faces. In the next phase, SIFT descriptors in the selected regions are concatenated to form feature vectors. Then these feature vectors are fed into linear discriminant analysis (LDA) for face recognition. In this study, faces are segmented into 47 regions and the descriptors in one or more regions are concatenated and their effect on face recognition was investigated. Face recognition tests were conducted by using the FRGC v2.0 face database.
Advances in hardware and pattern recognition techniques, along with the widespread utilization of... more Advances in hardware and pattern recognition techniques, along with the widespread utilization of remote sensing satellites, have urged the development of automatic target detection systems in satellite images. Automatic detection of airports is particularly essential, due to the strategic importance of these targets. In this paper, a runway detection method using a segmentation process based on textural properties is proposed
2012 IEEE International Geoscience and Remote Sensing Symposium, 2012
ABSTRACT The aim of this study is to automatically differentiate railroads from roads obtained by... more ABSTRACT The aim of this study is to automatically differentiate railroads from roads obtained by a general road extraction algorithm. In high resolution aerial images, the sleepers between railroad tracks depict a distinctive texture from other elongated objects. By employing this additional cue, a new Fourier based feature descriptor is developed to distinguish railroad objects in aerial images. After local Fourier features calculation, the periodic railroad texture is locally detected in the frequency domain.
2008 IEEE 16th Signal Processing, Communication and Applications Conference, 2008
... Effects of 3D Registration on Subspace Based Face Recognition Methods Bülent Üstün, Uğur Halı... more ... Effects of 3D Registration on Subspace Based Face Recognition Methods Bülent Üstün, Uğur Halıcı, İlkay Ulusoy, Tolga İnan ... Arkaplanda ve özellikle saç, kirpik, kaş, burun altı, çene altı gibi bölgelerde bazı 2B görüntü piksellerine karşılık 3B değer bulunmamaktadır. ...
2006 IEEE 14th Signal Processing and Communications Applications
3D face modeling based on real images is one of the important subjects of Computer Vision that is... more 3D face modeling based on real images is one of the important subjects of Computer Vision that is studied recently. In this paper the study that we conducted in our Computer Vision and Intelligent Systems Research Laboratory on 3D face model generation using uncalibrated multiple still images is explained
2012 20th Signal Processing and Communications Applications Conference (SIU), 2012
ABSTRACT In this study, an automatic and rule based approach which extracts bridges over water in... more ABSTRACT In this study, an automatic and rule based approach which extracts bridges over water in satellite images is developed. To find the bridges in the image, possible water regions are first found using thresholding and clustering methods. Candidate bridge regions are then extracted by applying morphological operations to the water mask. Also the algorithm proposes an optional verification step that uses geometric constraints and orientation information to increase precision values. The proposed algorithm is tested on different satellite images and the quantitative and visual results that were obtained showed that the algorithm is effective on extracting bridges of different sizes.
In sparse stereo [6], [12], [13], [14], [18], distinctive image features are extracted and corres... more In sparse stereo [6], [12], [13], [14], [18], distinctive image features are extracted and corresponding pairs are matched using a feature-based criteria. To be successful for stereo applications, local features must be robust to image deformations such as noise, rotation, scale and brightness changes. Jenkin, Jepson and Fleet [7], [8], [9] and Sanger [19] describe promising methods based on the output phase behavior of band-pass Gabor filters. Recently, Carneiro and Jepson show that the phase information provided by steerable filters is often locally stable with respect ...
2009 IEEE 17th Signal Processing and Communications Applications Conference, 2009
ABSTRACT The data acquired by 3D face scanners have distortions such as spikes, holes and noise. ... more ABSTRACT The data acquired by 3D face scanners have distortions such as spikes, holes and noise. Enhancement of 3D face data by removing these distortions while keeping the face features is important for the applications using these data. In this study, thresholding is used for removing spikes, thresholding together with face symmetry is used for hole filling and bilateral filtering is used for smoothing and satisfactory results are obtained on FRGC 3D face data.
[Proceedings 1992] IJCNN International Joint Conference on Neural Networks
A neural network approach for playing the game tic-tac-toe is introduced. The problem is consider... more A neural network approach for playing the game tic-tac-toe is introduced. The problem is considered as a combinatorial optimization problem aiming to maximize the value of a heuristic evaluation function. The proposed design guarantees a feasible solution, including in the cases where a winning move is never missed and a losing position is always prevented, if possible. The design has
In this paper we describe an algorithm for object recognition and cognitive map formation using s... more In this paper we describe an algorithm for object recognition and cognitive map formation using stereo image data in a D virtual world where 3D objects and a robot with stereo imaging system are simulated. Stereo imaging system is simulated so that the actual human visual system properties such as focusing, accommodation, field of view are parameterized. Only the stereo images obtained from this world are supplied to the virtual robot (agent). By applying our disparity algorithm on stereo image pairs, depth map for the current view is obtained. Using the depth information for the current view, a cognitive map of the environment is updated gradually while the virtual agent is exploring the environment. The agent explores its environment in an intelligent way using the current view and environmental map information obtained up to date. Also, during exploration if a new object is observed, from its view from different directions, it is labeled with its shape such as sphere, cylinder, cone
Recent studies have shown that lung cancer screening using annual low-dose computed tomography (C... more Recent studies have shown that lung cancer screening using annual low-dose computed tomography (CT) reduces lung cancer mortality by 20% compared to traditional chest radiography. Therefore, CT lung screening has started to be used widely all across the world. However, analyzing these images is a serious burden for radiologists. In this study, we propose a novel and simple framework that analyzes CT lung screenings using convolutional neural networks (CNNs) and reduces false positives. Our framework shows that even non-complex architectures are very powerful to classify 3D nodule data when compared to traditional methods. We also use different fusions in order to show their power and effect on the overall score. 3D CNNs are preferred over 2D CNNs because data are in 3D, and 2D convolutional operations may result in information loss. Mini-batch is used in order to overcome class-imbalance. Proposed framework has been validated according to the LUNA16 challenge evaluation and got scor...
Visual face tracking is one of the most important tasks in video surveillance systems. However, d... more Visual face tracking is one of the most important tasks in video surveillance systems. However, due to the variations in pose, scale, expression, and illumination it is considered to be a difficult task. Recent studies show that deep learning methods have a significant potential in object tracking tasks and adaptive feature selection methods can boost their performance. Motivated by these, we propose an end-to-end attentive deep learning based tracker, that is build on top of the state-of-the-art GOTURN tracker, for the task of real-time visual face tracking in video surveillance. Our method outperforms the state-of-the-art GOTURN and IVT trackers by very large margins and it achieves speeds that are very far beyond the requirements of real-time tracking. Additionally, to overcome the scarce data problem in visual face tracking, we also provide bounding box annotations for the G1 and G2 sets of ChokePoint dataset and make it suitable for further studies in face tracking under survei...
2016 20th National Biomedical Engineering Meeting (BIYOMUT), 2016
Laboratory mice are frequently used in biomedical studies. Facial expressions of mice provides im... more Laboratory mice are frequently used in biomedical studies. Facial expressions of mice provides important data about various issues. For this reason real time tracking of mice provide output to both researcher and software that operate on face image directly. Since body and face of laboratory mice is the same color and mice moves fast, tracking of face of mice is a challenging task. In recent years, methods that uses artificial neural networks provide effective solutions to problems such as classification, decision making and object recognition due to their ability to abstract training from data. In this study, a method based on deep learning is proposed for real time tracking of face of mice and successful results are obtained. Our studies are still going on in order to improve our results obtained using a limited dataset.
In this study, the effect of face regions on local shape descriptor based 3D face recognition was... more In this study, the effect of face regions on local shape descriptor based 3D face recognition was investigated. Our approach starts with calculation of the SIFT descriptors on the shape maps of the 3D faces. In the next phase, SIFT descriptors in the selected regions are concatenated to form feature vectors. Then these feature vectors are fed into linear discriminant analysis (LDA) for face recognition. In this study, faces are segmented into 47 regions and the descriptors in one or more regions are concatenated and their effect on face recognition was investigated. Face recognition tests were conducted by using the FRGC v2.0 face database.
Advances in hardware and pattern recognition techniques, along with the widespread utilization of... more Advances in hardware and pattern recognition techniques, along with the widespread utilization of remote sensing satellites, have urged the development of automatic target detection systems in satellite images. Automatic detection of airports is particularly essential, due to the strategic importance of these targets. In this paper, a runway detection method using a segmentation process based on textural properties is proposed
2012 IEEE International Geoscience and Remote Sensing Symposium, 2012
ABSTRACT The aim of this study is to automatically differentiate railroads from roads obtained by... more ABSTRACT The aim of this study is to automatically differentiate railroads from roads obtained by a general road extraction algorithm. In high resolution aerial images, the sleepers between railroad tracks depict a distinctive texture from other elongated objects. By employing this additional cue, a new Fourier based feature descriptor is developed to distinguish railroad objects in aerial images. After local Fourier features calculation, the periodic railroad texture is locally detected in the frequency domain.
2008 IEEE 16th Signal Processing, Communication and Applications Conference, 2008
... Effects of 3D Registration on Subspace Based Face Recognition Methods Bülent Üstün, Uğur Halı... more ... Effects of 3D Registration on Subspace Based Face Recognition Methods Bülent Üstün, Uğur Halıcı, İlkay Ulusoy, Tolga İnan ... Arkaplanda ve özellikle saç, kirpik, kaş, burun altı, çene altı gibi bölgelerde bazı 2B görüntü piksellerine karşılık 3B değer bulunmamaktadır. ...
2006 IEEE 14th Signal Processing and Communications Applications
3D face modeling based on real images is one of the important subjects of Computer Vision that is... more 3D face modeling based on real images is one of the important subjects of Computer Vision that is studied recently. In this paper the study that we conducted in our Computer Vision and Intelligent Systems Research Laboratory on 3D face model generation using uncalibrated multiple still images is explained
2012 20th Signal Processing and Communications Applications Conference (SIU), 2012
ABSTRACT In this study, an automatic and rule based approach which extracts bridges over water in... more ABSTRACT In this study, an automatic and rule based approach which extracts bridges over water in satellite images is developed. To find the bridges in the image, possible water regions are first found using thresholding and clustering methods. Candidate bridge regions are then extracted by applying morphological operations to the water mask. Also the algorithm proposes an optional verification step that uses geometric constraints and orientation information to increase precision values. The proposed algorithm is tested on different satellite images and the quantitative and visual results that were obtained showed that the algorithm is effective on extracting bridges of different sizes.
In sparse stereo [6], [12], [13], [14], [18], distinctive image features are extracted and corres... more In sparse stereo [6], [12], [13], [14], [18], distinctive image features are extracted and corresponding pairs are matched using a feature-based criteria. To be successful for stereo applications, local features must be robust to image deformations such as noise, rotation, scale and brightness changes. Jenkin, Jepson and Fleet [7], [8], [9] and Sanger [19] describe promising methods based on the output phase behavior of band-pass Gabor filters. Recently, Carneiro and Jepson show that the phase information provided by steerable filters is often locally stable with respect ...
2009 IEEE 17th Signal Processing and Communications Applications Conference, 2009
ABSTRACT The data acquired by 3D face scanners have distortions such as spikes, holes and noise. ... more ABSTRACT The data acquired by 3D face scanners have distortions such as spikes, holes and noise. Enhancement of 3D face data by removing these distortions while keeping the face features is important for the applications using these data. In this study, thresholding is used for removing spikes, thresholding together with face symmetry is used for hole filling and bilateral filtering is used for smoothing and satisfactory results are obtained on FRGC 3D face data.
[Proceedings 1992] IJCNN International Joint Conference on Neural Networks
A neural network approach for playing the game tic-tac-toe is introduced. The problem is consider... more A neural network approach for playing the game tic-tac-toe is introduced. The problem is considered as a combinatorial optimization problem aiming to maximize the value of a heuristic evaluation function. The proposed design guarantees a feasible solution, including in the cases where a winning move is never missed and a losing position is always prevented, if possible. The design has
In this paper we describe an algorithm for object recognition and cognitive map formation using s... more In this paper we describe an algorithm for object recognition and cognitive map formation using stereo image data in a D virtual world where 3D objects and a robot with stereo imaging system are simulated. Stereo imaging system is simulated so that the actual human visual system properties such as focusing, accommodation, field of view are parameterized. Only the stereo images obtained from this world are supplied to the virtual robot (agent). By applying our disparity algorithm on stereo image pairs, depth map for the current view is obtained. Using the depth information for the current view, a cognitive map of the environment is updated gradually while the virtual agent is exploring the environment. The agent explores its environment in an intelligent way using the current view and environmental map information obtained up to date. Also, during exploration if a new object is observed, from its view from different directions, it is labeled with its shape such as sphere, cylinder, cone
Uploads
Papers by Ugur Halici