Emre Turgay

Directionally adaptive super-resolution

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 the other edge-preserving methods, the proposed algorithm uses gradient direction in addition to the gradient amplitude for optimum regularization. The method comprises a gradient amplitude and direction estimation stage where a gradient direction map is obtained. This map guides the SR reconstruction stage through iterations. Three variations of the proposed method are compared against other edge-preserving super resolution methods. PSNR (Peak signal-to-noise-ratio), SSIM (Structural similarity index measure) values, and illustrations show that the proposed method has better performance especially on image pixel values where a strong gradient is present.

Context based super resolution image reconstruction

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. Texture segmentation and gradient map estimation are done prior to the reconstruction stage. Gradient direction is used for optimal noise reduction along the edges for non-textured regions. On the other hand, regularization term is cancelled for textured regions so that the resultant method reduces to maximum likelihood (ML) solution. It is demonstrated on Brodatz Texture Database that ML solution gives the best PSNR values on textures compared to the regularized SR methods in the literature. Experimental results show that the proposed hybrid method has superior performance in terms of Peak Signal-to-Noise-Ratio (PSNR), Structural Similarity Index Measure (SSIM) compared the SR methods in the literature.

Texture and edge preserving multiframe super-resolution

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-a-posteriori 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.

FPGA based real-time efficient histogram equalization

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 FPGAs and DSPs, and their output video quality on current systems. In these implementation methods histogram transformation function is described by 2, 16 or 256 pieces. DSP work load is totally released on 256 piece method. The number of configurable logic blocks used on FPGA is decreased without any loss on speed. By implementing 256 pieces method the complexity of the method is decreased and an increase in the quality of the output images obtained. The results are presented on scanning array thermal cameras.

Autofocus method in thermal cameras based on image histogram

In this paper, a new histogram based auto-focusing method for thermal cameras is proposed. This proposed method is realized by FPGA (Field Programmable Gate Array) and DSP (Digital Signal Processor) working together and simultaneously. HF (High Frequency) component, obtained from real-time image flow by FPGA and DSP is used for auto-focusing process. Proposed method is able to determine the focus direction from the HF component produced in the process of histogram equalization by FPGA, unlike Fourier transform and pixel differenve based methods in the literature. With this superiority, proposed method requires no extra calculation for thermal cameras for which histogram equalization is necessary. Analysis show that proposed method is successful on the simulations and scanning thermal cameras.

Color super resolution in HSV domain

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-to-noise-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 channels is sufficient to increase the real-time performance of the image processing systems.

Texture preserving multi frame super resolution with spatially varying image prior

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.

Texture preserving super resolution in thermal images

This paper proposes a new super-resolution (SR) image reconstruction method targeting edges and textured regions in thermal images. Proposed two stage method runs two Bayesian SR estimators at the first stage. These estimators are; maximum likelihood method and maximum a-posteriori method with Tikhonov type regularization. The piksel-to-piksel difference image of these two estimates is an high frequency (HF) image including observation noise, process noise, edges and textures (that are smoothed out by regularizers). In the second stage of the proposed method, the difference image is post-processed to extract edge and texture information while eliminating noise. The proposed method uses a Gabor filter family to analyze this difference image at various frequencies and directions. The strong frequency components are restored and added to the MAP estimate to obtain the final image. The proposed methods are validated through simulations on several textured surfaces from Brodatz data base. Peak-signal-to-noise ratio (PSNR) measures and illustrations clearly shows the success of the proposed method. The Real experiments on uncooled thermal cameras are also conducted to compare the methods to classical SR methods known to literature.

Direction Adaptive Super-Resolution Imaging

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-noise-ratio (PSNR) measures and illustrations clearly show that the proposed method is successful especially on edge structures in images.