Image Interpolation by Super-Resolution

Science and Technology Development Journal, 2013

This paper presents an efficient method for video super-resolution (SR) based on two main ideals: Firstly, input video frames can be separated into two components, nontexturing image and texturing image. Then each component image is applied to a compatible interpolation method to improve the quality of high-resolution (HR) reconstructed frame. Secondly, based on the approach that border regions of image details are the most lossy information regions from the sampling process. Therefore, a task of compensation interpolation is essential to increase the quality of the reconstructed HR images. From these discussions, we proposed an efficient method for video SR by combining the spatial interpolation in different texturing regions and the sampling compensation interpolation to improve the quality of video super-resolution. Our results shown that, the quality of HR frames, reconstructed by the proposed method, is better than that of other methods, , and in recently. The significant contr...

Multimedia Tools and Applications, 2012

This paper proposes a cost-effective and edge-directed image super-resolution scheme. Image super-resolution (image magnification) is an enthusiastic research area and is desired in a variety of applications. The basic idea of the proposed scheme is based on the concept of multi-kernel approach. Various stencils have been defined on the basis of geometrical regularities. This set of stencils is associated with the set of kernels. The value of a re-sampling pixel is obtained by calculating the weighted average of the pixels in the selected kernel. The time complexity of the proposed scheme is as low as that of classical linear interpolation techniques, but the visual quality is more appealing because of the edgeorientation property. The experimental results and analysis show that proposed scheme provides a good combination of visual quality and time complexity.

This paper presents a new approach to digital image super-resolution (SR). Image SR is currently a very active area of research because it is used in various applications. The proposed technique uses Gaussian edge directed interpolation to determine the precise weights of the neighboring pixels. The standard deviation of the interpolation window determines the value of the sigma ‘σ’ for generating Gaussian kernels. Therefore, the proposed scheme adaptively applies different Gaussian kernels according to the computed standard deviation of the interpolation window. Laplacian is applied to the image generated by the Gaussian kernels to enhance the visual quality of the output image. It has the significant benefit of being isotropic i.e. invariant to rotation. These features of being isotropic not only resemble human visual perception but also respond to intensity variations equally in all directions for any kind of kernel. It highlights the discontinuities of high frequencies in the image generated by the Gaussian kernel and deemphasizes the regions with slowly varying luminance levels. It also recovers the background missing features while preserving the sharpness of the output image. The proposed scheme preserves geometrical regularities across the boundaries and smoothes intensities inside the high frequencies. It also maintains the textures inside geometrical regularities. Therefore, high resolution (HR) images produced by the proposed scheme contain intensity information very close to the original details of the low-resolution (LR) image i.e. edges, smoothness and texture information. Various evaluation metrics have been applied to compute the validity of the proposed technique. Extensive experimental comparisons with state-of-the-art zooming schemes validate the claim of the proposed technique of being superior. It produces high quality at the cost of low time complexity.

IEEE Transactions on Image Processing, 2006

In this paper, we propose an image super-resolution (resolution enhancement) algorithm that takes into account inaccurate estimates of the registration parameters and the point spread function. These inaccurate estimates, along with the additive Gaussian noise in the low-resolution (LR) image sequence, result in different noise level for each frame. In the proposed algorithm, the LR frames are adaptively weighted according to their reliability and the regularization parameter is simultaneously estimated. A translational motion model is assumed. The convergence property of the proposed algorithm is analyzed in detail. Our experimental results using both real and synthetic data show the effectiveness of the proposed algorithm.

Signal, Image and Video Processing, 2014

This paper presents a fast single-image superresolution approach that involves learning multiple adaptive interpolation kernels. Based on the assumptions that each high-resolution image patch can be sparsely represented by several simple image structures and that each structure can be assigned a suitable interpolation kernel, our approach consists of the following steps. First, we cluster the training image patches into several classes and train each class-specific interpolation kernel. Then, for each input lowresolution image patch, we select few suitable kernels of it to make up the final interpolation kernel. Since the proposed approach is mainly based on simple linear algebra computations, its efficiency can be guaranteed. And experimental comparisons with state-of-the-art super-resolution reconstruction algorithms on simulated and real-life examples can validate the performance of our proposed approach.

Computer Vision, 2009 IEEE 12th …, 2009

Methods for super-resolution can be broadly classified into two families of methods: (i) The classical multi-image super-resolution (combining images obtained at subpixel misalignments), and (ii) Example-Based super-resolution (learning correspondence between low and high resolution image patches from a database). In this paper we propose a unified framework for combining these two families of methods. We further show how this combined approach can be applied to obtain super resolution from as little as a single image (with no database or prior examples). Our approach is based on the observation that patches in a natural image tend to redundantly recur many times inside the image, both within the same scale, as well as across different scales. Recurrence of patches within the same image scale (at subpixel misalignments) gives rise to the classical super-resolution, whereas recurrence of patches across different scales of the same image gives rise to example-based super-resolution. Our approach attempts to recover at each pixel its best possible resolution increase based on its patch redundancy within and across scales.

Signal Processing, 2010

Multi-frame image super-resolution (SR) aims to utilize information from a set of lowresolution (LR) images to compose a high-resolution (HR) one. As it is desirable or essential in many real applications, recent years have witnessed the growing interest in the problem of multi-frame SR reconstruction. This set of algorithms commonly utilizes a linear observation model to construct the relationship between the recorded LR images to the unknown reconstructed HR image estimates. Recently, regularizationbased schemes have been demonstrated to be effective because SR reconstruction is actually an ill-posed problem. Working within this promising framework, this paper first proposes two new regularization items, termed as locally adaptive bilateral total variation and consistency of gradients, to keep edges and flat regions, which are implicitly described in LR images, sharp and smooth, respectively. Thereafter, the combination of the proposed regularization items is superior to existing regularization items because it considers both edges and flat regions while existing ones consider only edges. Thorough experimental results show the effectiveness of the new algorithm for SR reconstruction.

IEEE Transactions on Image Processing, 2014

This paper presents a novel example-based singleimage super-resolution (SR) procedure, that upscales to highresolution (HR) a given low-resolution (LR) input image without relying on an external dictionary of image examples. The dictionary instead is built from the LR input image itself, by generating a "double pyramid" of recursively scaled, and subsequently interpolated, images, from which self-examples are extracted. The upscaling procedure is multi-pass, i.e. the output image is constructed by means of gradual increases, and consists in learning special linear mapping functions on this double pyramid, as many as the number of patches in the current image to upscale. More precisely, for each LR patch, similar self-examples are found, and, thanks to them, a linear function is learned to directly map it into its HR version. Iterative back projection is also employed to ensure consistency at each pass of the procedure. Extensive experiments and comparisons with other state-of-theart methods, based both on external and internal dictionaries, show that our algorithm can produce visually pleasant upscalings, with sharp edges and well reconstructed details. Moreover, when considering objective metrics like PSNR and SSIM, our method turns out to give the best performance.

ArXiv.org, 2018

We propose a novel architecture that learns an end-to-end mapping function to improve the spatial resolution of the input natural images. The model is unique in forming a nonlinear combination of three traditional interpolation techniques using the convolutional neural network. Another proposed architecture uses a skip connection with nearest neighbor interpolation, achieving almost similar results. The architectures have been carefully designed to ensure that the reconstructed images lie precisely in the manifold of high-resolution images, thereby preserving the high-frequency components with fine details. We have compared with the state of the art and recent deep learning based natural image super-resolution techniques and found that our methods are able to preserve the sharp details in the image, while also obtaining comparable or better PSNR than them. Since our methods use only traditional interpolations and a shallow CNN with less number of smaller filters, the computational cost is kept low. We have reported the results of two proposed architectures on five standard datasets, for an upscale factor of 2. Our methods generalize well in most cases, which is evident from the better results obtained with increasingly complex datasets. For 4-times upscaling, we have designed similar architectures for comparing with other methods.

IEEE Access

Super-Resolution (SR) was an important research topic, and SR methods based on Convolutional Neural Network (CNN) confirmed its groundbreaking performance. However, notably implementing the CNN model into resource-limited hardware devices is a great challenge. Therefore, we present a hardware-friendly and low-cost interpolation for Multi-Magnification SR image reconstruction. We follow our previous work, which is a learning-based interpolation (LCDI) with a self-defined classifier of image texture, and extend its original ×2 architecture to ×3 and ×4 architecture. Besides, the required pre-trained weights are reduced by the fusion scheme. Experimentally, the proposed method has only 75% lower pre-trained weights than LCDI. Compared to the related work OLM-SI (One linear learning mapping-SI), the run-time and quantity of pre-trained weights of the ×2 proposed method are at least 90% lower. Compared to CNN-based SR methods, the proposed method loses a little lower performance, but the evaluation of computational cost is much lower. In conclusion, the proposed method is cost-effective and a practical solution for resource-limited hardware and device.