Emre Turgay - Academia.edu (original) (raw)
Papers by Emre Turgay
... TEXTURE PRESERVING SUPER RESOLUTION IN THERMAL IMAGES Emre Turgay1,2, Gözde Bozda˘gı Akar2, .... more ... TEXTURE PRESERVING SUPER RESOLUTION IN THERMAL IMAGES Emre Turgay1,2, Gözde Bozda˘gı Akar2, ... Bu görüntüde gökyüzünde detay olmadı˘gı için MAP çözüm tercih edilirken, özellikle elektrik dire˘gi gövdesindeki desen temiz bir sekilde ortaya çıkartılmıstır. ...
Iet Image Processing, Sep 1, 2014
Super-resolution (SR) image reconstruction refers to methods where a higher resolution image is r... more Super-resolution (SR) image reconstruction refers to methods where a higher resolution image is reconstructed using a set of overlapping aliased low-resolution observations of the same scene. Although edge preservation has been a widely explored topic in SR literature, texture-specific regularisation has recently gained interest. In this study, texture-specific regularisation is handled as a post-processing step. A two stage method is proposed, comprising multiple SR reconstructions with different regularisation parameters followed by a restoration step for preserving edges and textures. In the first stage, two maximum-aposteriori estimators with two different amounts of regularisation are employed. In the second stage, pixel-to-pixel difference between these two estimates is post-processed to restore edges and textures. Frequency selective characteristics of discrete cosine transform and Gabor filters are utilised in the post-processing step. Experiments on synthetically generated images and real experiments demonstrate that the proposed methods give better results compared with the state-of-the-art SR methods especially on textures and edges.
ABSTRACT This paper proposes a new maximum a posteriori (MAP) based super-resolution (SR) image r... more ABSTRACT This paper proposes a new maximum a posteriori (MAP) based super-resolution (SR) image reconstruction method targeting edges and textures in images. Unlike conventional MAP based SR image reconstruction methods a spatially varying image prior is employed which is updated according to the frequency content of the reconstructed image at each iteration at different locations. Two alternative methods based on discrete cosine transforms (DCT) and Gabor filters are proposed for determining the image prior. The proposed method is validated through simulations and real experiments which clearly demonstrates significant visual improvements especially on edges and textures compared to state-of-the-art SR methods.
Abstract In this paper a novel direction adaptive super-resolution (SR) image reconstruction meth... more Abstract In this paper a novel direction adaptive super-resolution (SR) image reconstruction method is proposed. The proposed maximum a-posteriori (MAP) based estimator uses gradient direction for optimal noise reduction while preserving the edges. Compared to ...
Abstract In this paper a context based super-resolution (SR) image reconstruction method is propo... more Abstract In this paper a context based super-resolution (SR) image reconstruction method is proposed. The proposed maximum a-posteriori (MAP) based estimator identifies local gradients and textures for selecting the optimal SR method for the region of interest. ...
In this paper a novel edge-preserving super-resolution (SR) image reconstruction method is propos... more In this paper a novel edge-preserving super-resolution (SR) image reconstruction method is proposed. The proposed maximum a-posteriori (MAP) based estimator uses gradient direction and amount for optimal noise reduction while preserving the edges. Compared to the other edge-preserving methods, the proposed algorithm uses the gradient direction for optimum regularization. The proposed method estimates gradient amplitude and direction at each iteration. This gradient map guides the SR reconstruction stage through iterations. Proposed method is compared against other traditional super resolution methods. Peak-signal-to-noiseratio (PSNR) measures and illustrations clearly show that the proposed method is successful especially on edge structures in images. 1. Giri
In this paper, we present the discovery of new haptic illusions, and a method for creating furthe... more In this paper, we present the discovery of new haptic illusions, and a method for creating further haptic illusions. Pure haptic illusions were created by mixing force cues with geometric cues to make people feel shapes that differ from the actual shape of the object. In particular, we found that an area that feels harder to move through and easier to move out of can be interpreted as a region of high curvature. These haptic illusions were implemented into user experiments with a haptic interface. Preliminary user tests have shown that this method can be effective at masking true object geometry by making a circle feel like an ellipse and a straight line feel like a curved line.
T h i s thesis describes a system to measure the mechanical properties of soft tissue. T h e mech... more T h i s thesis describes a system to measure the mechanical properties of soft tissue. T h e mechanica l properties depend on the type of tissue (e.g. fat, muscle, b lood) and the presence of disease or pathology. For example, i t has been known that many cancers, such as ca rc inoma of the breast a n d the prostate, appear as ha rd nodules. M a n u a l pa lpa t ion is wide ly used for detect ing these nodules but detection by pa lpa t ion is restr icted to large tumors that reside relat ively close to an accessible surface. Cur ren t imaging devices such as computed tomography, magnetic resonance imaging and u l t rasound are not d i rec t ly capable of measuring the mechanical properties of tissue. A n emerging research topic, called elastography, a ims to produce a new type of image that depicts mechanical properties of tissue. T h e basic pr inc ip le is to excite mo t ion i n a tissue, record the mot ion w i t h an imaging device, and then estimate the mechanical properties fro...
In this paper, a new color super-resolution (SR) image reconstruction method for HSV domain is pr... more In this paper, a new color super-resolution (SR) image reconstruction method for HSV domain is presented. Applying classical SR algorithms to each channel of the HSV video causes an artifact due to the cyclic nature of the hue channel. Proposed method solves this problem without converting the HSV video to RGB video. The proposed method reconstructs the saturation and value channels using maximum aposteriori based SR algorithm. For the hue channel the problematic pixels causing the artifacts are masked and a nearest neighbor interpolation is applied. The proposed method is compared to the classical approach where HSV video is converted to RGB domain and each channel is processed by a MAP based SR algorithm. Peak-signal-tonoise-ratio (PSNR) measures show that the proposed approach does not lower the PSNR values. In this paper, it has also been shown that increasing the resolution of the hue channel does not affect the detail perception. Hence applying SR methods to only S and V chann...
2012 19th IEEE International Conference on Image Processing, 2012
ABSTRACT This paper proposes a new maximum a posteriori (MAP) based super-resolution (SR) image r... more ABSTRACT This paper proposes a new maximum a posteriori (MAP) based super-resolution (SR) image reconstruction method targeting edges and textures in images. Unlike conventional MAP based SR image reconstruction methods a spatially varying image prior is employed which is updated according to the frequency content of the reconstructed image at each iteration at different locations. Two alternative methods based on discrete cosine transforms (DCT) and Gabor filters are proposed for determining the image prior. The proposed method is validated through simulations and real experiments which clearly demonstrates significant visual improvements especially on edges and textures compared to state-of-the-art SR methods.
2011 IEEE 19th Signal Processing and Communications Applications Conference (SIU), 2011
2009 International Workshop on Local and Non-Local Approximation in Image Processing, 2009
Abstract In this paper a context based super-resolution (SR) image reconstruction method is propo... more Abstract In this paper a context based super-resolution (SR) image reconstruction method is proposed. The proposed maximum a-posteriori (MAP) based estimator identifies local gradients and textures for selecting the optimal SR method for the region of interest. ...
2011 IEEE 19th Signal Processing and Communications Applications Conference (SIU), 2011
Abstract In this paper, a new color super-resolution (SR) image reconstruction method for HSV dom... more Abstract In this paper, a new color super-resolution (SR) image reconstruction method for HSV domain is presented. Applying classical SR algorithms to each channel of the HSV video causes an artifact due to the cyclic nature of the hue channel. Proposed method ...
2011 IEEE 19th Signal Processing and Communications Applications Conference (SIU), 2011
... TEXTURE PRESERVING SUPER RESOLUTION IN THERMAL IMAGES Emre Turgay1,2, Gözde Bozda˘gı Akar2, .... more ... TEXTURE PRESERVING SUPER RESOLUTION IN THERMAL IMAGES Emre Turgay1,2, Gözde Bozda˘gı Akar2, ... Bu görüntüde gökyüzünde detay olmadı˘gı için MAP çözüm tercih edilirken, özellikle elektrik dire˘gi gövdesindeki desen temiz bir sekilde ortaya çıkartılmıstır. ...
2009 16th IEEE International Conference on Image Processing (ICIP), 2009
ABSTRACT
2009 IEEE 17th Signal Processing and Communications Applications Conference, 2009
Page 1. Yön Uyarlamalı Süper-Çözünürlük Görüntüleme Yöntemi Direction Adaptive Super-Resolution I... more Page 1. Yön Uyarlamalı Süper-Çözünürlük Görüntüleme Yöntemi Direction Adaptive Super-Resolution Imaging Emre Turgay 1,2 , Gözde Bozda÷ı Akar 2 1. Görüntü İşleme Müdürlüğü, Mikroelektronik Güdüm ve Elektro-Optik Grubu, ASELSAN eturgay@mgeo.aselsan.com.tr ...
IET Image Processing, 2014
Super-resolution (SR) image reconstruction refers to methods where a higher resolution image is r... more Super-resolution (SR) image reconstruction refers to methods where a higher resolution image is reconstructed using a set of overlapping aliased low-resolution observations of the same scene. Although edge preservation has been a widely explored topic in SR literature, texture-specific regularisation has recently gained interest. In this study, texture-specific regularisation is handled as a post-processing step. A two stage method is proposed, comprising multiple SR reconstructions with different regularisation parameters followed by a restoration step for preserving edges and textures. In the first stage, two maximum-aposteriori estimators with two different amounts of regularisation are employed. In the second stage, pixel-to-pixel difference between these two estimates is post-processed to restore edges and textures. Frequency selective characteristics of discrete cosine transform and Gabor filters are utilised in the post-processing step. Experiments on synthetically generated images and real experiments demonstrate that the proposed methods give better results compared with the state-of-the-art SR methods especially on textures and edges.
Ultrasound in Medicine & Biology, 2006
A new elastography method is presented that images the mechanical properties of soft tissue. The ... more A new elastography method is presented that images the mechanical properties of soft tissue. The tissue is externally vibrated over a range of frequencies simultaneously and the resulting displacement is recorded at multiple locations and time instants with a sequence of ultrasound images. Two methods are proposed for estimating the mechanical properties from the recorded data. In the first method, equations of motion are written for a tissue model of masses, springs and dampers. The equations are solved to identify the model parameters and construct an image. The second method performs a transfer function analysis of the tissue motion. The tissue properties are identified from the shape the of transfer function. Simulations show good performance of the new method compared with static elastography. Initial experimental results from homogeneous and layered tissue phantoms also demonstrate the ability quantitatively to image tissue stiffness. Preliminary results are obtained for viscosity and density estimation. Further work is needed to extend the formulation to 3-D and improve robustness of the viscosity and density estimates.
In this paper, we present the discovery of new haptic illusions, and a method for creating furthe... more In this paper, we present the discovery of new haptic illusions, and a method for creating further haptic illusions. Pure haptic illusions were created by mixing force cues with geometric cues to make people feel shapes that differ from the actual shape of the object. In particular, we found that an area that feels harder to move through and easier to move out of can be interpreted as a region of high curvature. These haptic illusions were implemented into user experiments with a haptic interface. Preliminary user tests have shown that this method can be effective at masking true object geometry by making a circle feel like an ellipse and a straight line feel like a curved line.
2009 IEEE 17th Signal Processing and Communications Applications Conference, 2009
Histogram equalization method is implemented by field programmable gate arrays (FPGA) and digital... more Histogram equalization method is implemented by field programmable gate arrays (FPGA) and digital signal processors (DSP) together on thermal cameras. In this paper, we discussed work load of different histogram equalization implementations on FPGA's and DSP's, and their output video quality on current systems. In these implementation methods histogram transformation function is described by 2, 16 or 256 pieces. DSP
... TEXTURE PRESERVING SUPER RESOLUTION IN THERMAL IMAGES Emre Turgay1,2, Gözde Bozda˘gı Akar2, .... more ... TEXTURE PRESERVING SUPER RESOLUTION IN THERMAL IMAGES Emre Turgay1,2, Gözde Bozda˘gı Akar2, ... Bu görüntüde gökyüzünde detay olmadı˘gı için MAP çözüm tercih edilirken, özellikle elektrik dire˘gi gövdesindeki desen temiz bir sekilde ortaya çıkartılmıstır. ...
Iet Image Processing, Sep 1, 2014
Super-resolution (SR) image reconstruction refers to methods where a higher resolution image is r... more Super-resolution (SR) image reconstruction refers to methods where a higher resolution image is reconstructed using a set of overlapping aliased low-resolution observations of the same scene. Although edge preservation has been a widely explored topic in SR literature, texture-specific regularisation has recently gained interest. In this study, texture-specific regularisation is handled as a post-processing step. A two stage method is proposed, comprising multiple SR reconstructions with different regularisation parameters followed by a restoration step for preserving edges and textures. In the first stage, two maximum-aposteriori estimators with two different amounts of regularisation are employed. In the second stage, pixel-to-pixel difference between these two estimates is post-processed to restore edges and textures. Frequency selective characteristics of discrete cosine transform and Gabor filters are utilised in the post-processing step. Experiments on synthetically generated images and real experiments demonstrate that the proposed methods give better results compared with the state-of-the-art SR methods especially on textures and edges.
ABSTRACT This paper proposes a new maximum a posteriori (MAP) based super-resolution (SR) image r... more ABSTRACT This paper proposes a new maximum a posteriori (MAP) based super-resolution (SR) image reconstruction method targeting edges and textures in images. Unlike conventional MAP based SR image reconstruction methods a spatially varying image prior is employed which is updated according to the frequency content of the reconstructed image at each iteration at different locations. Two alternative methods based on discrete cosine transforms (DCT) and Gabor filters are proposed for determining the image prior. The proposed method is validated through simulations and real experiments which clearly demonstrates significant visual improvements especially on edges and textures compared to state-of-the-art SR methods.
Abstract In this paper a novel direction adaptive super-resolution (SR) image reconstruction meth... more Abstract In this paper a novel direction adaptive super-resolution (SR) image reconstruction method is proposed. The proposed maximum a-posteriori (MAP) based estimator uses gradient direction for optimal noise reduction while preserving the edges. Compared to ...
Abstract In this paper a context based super-resolution (SR) image reconstruction method is propo... more Abstract In this paper a context based super-resolution (SR) image reconstruction method is proposed. The proposed maximum a-posteriori (MAP) based estimator identifies local gradients and textures for selecting the optimal SR method for the region of interest. ...
In this paper a novel edge-preserving super-resolution (SR) image reconstruction method is propos... more In this paper a novel edge-preserving super-resolution (SR) image reconstruction method is proposed. The proposed maximum a-posteriori (MAP) based estimator uses gradient direction and amount for optimal noise reduction while preserving the edges. Compared to the other edge-preserving methods, the proposed algorithm uses the gradient direction for optimum regularization. The proposed method estimates gradient amplitude and direction at each iteration. This gradient map guides the SR reconstruction stage through iterations. Proposed method is compared against other traditional super resolution methods. Peak-signal-to-noiseratio (PSNR) measures and illustrations clearly show that the proposed method is successful especially on edge structures in images. 1. Giri
In this paper, we present the discovery of new haptic illusions, and a method for creating furthe... more In this paper, we present the discovery of new haptic illusions, and a method for creating further haptic illusions. Pure haptic illusions were created by mixing force cues with geometric cues to make people feel shapes that differ from the actual shape of the object. In particular, we found that an area that feels harder to move through and easier to move out of can be interpreted as a region of high curvature. These haptic illusions were implemented into user experiments with a haptic interface. Preliminary user tests have shown that this method can be effective at masking true object geometry by making a circle feel like an ellipse and a straight line feel like a curved line.
T h i s thesis describes a system to measure the mechanical properties of soft tissue. T h e mech... more T h i s thesis describes a system to measure the mechanical properties of soft tissue. T h e mechanica l properties depend on the type of tissue (e.g. fat, muscle, b lood) and the presence of disease or pathology. For example, i t has been known that many cancers, such as ca rc inoma of the breast a n d the prostate, appear as ha rd nodules. M a n u a l pa lpa t ion is wide ly used for detect ing these nodules but detection by pa lpa t ion is restr icted to large tumors that reside relat ively close to an accessible surface. Cur ren t imaging devices such as computed tomography, magnetic resonance imaging and u l t rasound are not d i rec t ly capable of measuring the mechanical properties of tissue. A n emerging research topic, called elastography, a ims to produce a new type of image that depicts mechanical properties of tissue. T h e basic pr inc ip le is to excite mo t ion i n a tissue, record the mot ion w i t h an imaging device, and then estimate the mechanical properties fro...
In this paper, a new color super-resolution (SR) image reconstruction method for HSV domain is pr... more In this paper, a new color super-resolution (SR) image reconstruction method for HSV domain is presented. Applying classical SR algorithms to each channel of the HSV video causes an artifact due to the cyclic nature of the hue channel. Proposed method solves this problem without converting the HSV video to RGB video. The proposed method reconstructs the saturation and value channels using maximum aposteriori based SR algorithm. For the hue channel the problematic pixels causing the artifacts are masked and a nearest neighbor interpolation is applied. The proposed method is compared to the classical approach where HSV video is converted to RGB domain and each channel is processed by a MAP based SR algorithm. Peak-signal-tonoise-ratio (PSNR) measures show that the proposed approach does not lower the PSNR values. In this paper, it has also been shown that increasing the resolution of the hue channel does not affect the detail perception. Hence applying SR methods to only S and V chann...
2012 19th IEEE International Conference on Image Processing, 2012
ABSTRACT This paper proposes a new maximum a posteriori (MAP) based super-resolution (SR) image r... more ABSTRACT This paper proposes a new maximum a posteriori (MAP) based super-resolution (SR) image reconstruction method targeting edges and textures in images. Unlike conventional MAP based SR image reconstruction methods a spatially varying image prior is employed which is updated according to the frequency content of the reconstructed image at each iteration at different locations. Two alternative methods based on discrete cosine transforms (DCT) and Gabor filters are proposed for determining the image prior. The proposed method is validated through simulations and real experiments which clearly demonstrates significant visual improvements especially on edges and textures compared to state-of-the-art SR methods.
2011 IEEE 19th Signal Processing and Communications Applications Conference (SIU), 2011
2009 International Workshop on Local and Non-Local Approximation in Image Processing, 2009
Abstract In this paper a context based super-resolution (SR) image reconstruction method is propo... more Abstract In this paper a context based super-resolution (SR) image reconstruction method is proposed. The proposed maximum a-posteriori (MAP) based estimator identifies local gradients and textures for selecting the optimal SR method for the region of interest. ...
2011 IEEE 19th Signal Processing and Communications Applications Conference (SIU), 2011
Abstract In this paper, a new color super-resolution (SR) image reconstruction method for HSV dom... more Abstract In this paper, a new color super-resolution (SR) image reconstruction method for HSV domain is presented. Applying classical SR algorithms to each channel of the HSV video causes an artifact due to the cyclic nature of the hue channel. Proposed method ...
2011 IEEE 19th Signal Processing and Communications Applications Conference (SIU), 2011
... TEXTURE PRESERVING SUPER RESOLUTION IN THERMAL IMAGES Emre Turgay1,2, Gözde Bozda˘gı Akar2, .... more ... TEXTURE PRESERVING SUPER RESOLUTION IN THERMAL IMAGES Emre Turgay1,2, Gözde Bozda˘gı Akar2, ... Bu görüntüde gökyüzünde detay olmadı˘gı için MAP çözüm tercih edilirken, özellikle elektrik dire˘gi gövdesindeki desen temiz bir sekilde ortaya çıkartılmıstır. ...
2009 16th IEEE International Conference on Image Processing (ICIP), 2009
ABSTRACT
2009 IEEE 17th Signal Processing and Communications Applications Conference, 2009
Page 1. Yön Uyarlamalı Süper-Çözünürlük Görüntüleme Yöntemi Direction Adaptive Super-Resolution I... more Page 1. Yön Uyarlamalı Süper-Çözünürlük Görüntüleme Yöntemi Direction Adaptive Super-Resolution Imaging Emre Turgay 1,2 , Gözde Bozda÷ı Akar 2 1. Görüntü İşleme Müdürlüğü, Mikroelektronik Güdüm ve Elektro-Optik Grubu, ASELSAN eturgay@mgeo.aselsan.com.tr ...
IET Image Processing, 2014
Super-resolution (SR) image reconstruction refers to methods where a higher resolution image is r... more Super-resolution (SR) image reconstruction refers to methods where a higher resolution image is reconstructed using a set of overlapping aliased low-resolution observations of the same scene. Although edge preservation has been a widely explored topic in SR literature, texture-specific regularisation has recently gained interest. In this study, texture-specific regularisation is handled as a post-processing step. A two stage method is proposed, comprising multiple SR reconstructions with different regularisation parameters followed by a restoration step for preserving edges and textures. In the first stage, two maximum-aposteriori estimators with two different amounts of regularisation are employed. In the second stage, pixel-to-pixel difference between these two estimates is post-processed to restore edges and textures. Frequency selective characteristics of discrete cosine transform and Gabor filters are utilised in the post-processing step. Experiments on synthetically generated images and real experiments demonstrate that the proposed methods give better results compared with the state-of-the-art SR methods especially on textures and edges.
Ultrasound in Medicine & Biology, 2006
A new elastography method is presented that images the mechanical properties of soft tissue. The ... more A new elastography method is presented that images the mechanical properties of soft tissue. The tissue is externally vibrated over a range of frequencies simultaneously and the resulting displacement is recorded at multiple locations and time instants with a sequence of ultrasound images. Two methods are proposed for estimating the mechanical properties from the recorded data. In the first method, equations of motion are written for a tissue model of masses, springs and dampers. The equations are solved to identify the model parameters and construct an image. The second method performs a transfer function analysis of the tissue motion. The tissue properties are identified from the shape the of transfer function. Simulations show good performance of the new method compared with static elastography. Initial experimental results from homogeneous and layered tissue phantoms also demonstrate the ability quantitatively to image tissue stiffness. Preliminary results are obtained for viscosity and density estimation. Further work is needed to extend the formulation to 3-D and improve robustness of the viscosity and density estimates.
In this paper, we present the discovery of new haptic illusions, and a method for creating furthe... more In this paper, we present the discovery of new haptic illusions, and a method for creating further haptic illusions. Pure haptic illusions were created by mixing force cues with geometric cues to make people feel shapes that differ from the actual shape of the object. In particular, we found that an area that feels harder to move through and easier to move out of can be interpreted as a region of high curvature. These haptic illusions were implemented into user experiments with a haptic interface. Preliminary user tests have shown that this method can be effective at masking true object geometry by making a circle feel like an ellipse and a straight line feel like a curved line.
2009 IEEE 17th Signal Processing and Communications Applications Conference, 2009
Histogram equalization method is implemented by field programmable gate arrays (FPGA) and digital... more Histogram equalization method is implemented by field programmable gate arrays (FPGA) and digital signal processors (DSP) together on thermal cameras. In this paper, we discussed work load of different histogram equalization implementations on FPGA's and DSP's, and their output video quality on current systems. In these implementation methods histogram transformation function is described by 2, 16 or 256 pieces. DSP