scispace - formally typeset
Search or ask a question
Proceedings ArticleDOI

A new edge directed interpolation algorithm using accurate estimation of edge directional covariance

20 May 2012-pp 1223-1226
TL;DR: Simulation result shows that the new interpolation algorithm significantly improves the subjective quality of the interpolated images compared with conventional linear interpolation and NEDI one and demonstrates the improvements of objective metrics such as PSNR, SSIM and WEA which are used for the accuracy estimation of directionality.
Abstract: This paper proposes an edge-directed interpolation algorithm to enhance the quality of natural images which are captured by low-resolution camera installed on car or CCTV. Based on the accurate estimation of edge directional covariance between low-resolution and high-resolution image, local covariance coefficients extracted from the low-resolution image has been adapted for the interpolation to obtain the high-resolution image. DCT (Discrete Cosine Transform) kernel function is used in order to reflect the multi-directional edge accurately without increasing of complexity. Simulation result shows that our new interpolation algorithm significantly improves the subjective quality of the interpolated images compared with conventional linear interpolation and NEDI one. It also demonstrates the improvements of objective metrics such as PSNR, SSIM(structural similarity index measurement) and WEA (Wiener filter coefficients Estimation Accuracy) which are used for the accuracy estimation of directionality.
Citations
More filters
Journal ArticleDOI
TL;DR: A learning-based image interpolation method based on weighted direct nonlinear regression that outperforms many state-of-the-art methods on the basis of effective and efficient performance and a refinement process to improve the interpolation performance with sharper edges and richer details.
Abstract: This paper proposes a learning-based image interpolation method based on weighted direct nonlinear regression. It attempts to learn the nonlinear relationship between the low-resolution patches and their corresponding high-resolution patches by using an external database of natural images. In the training phase, without being initialized to the same size as the high-resolution patches using bicubic interpolation, low-resolution patches are directly used for training, which will reduce the number of parameters. Dictionary learning combined with nearest neighbor searching based on the normalized correlation coefficient in the entire training set is proposed as a soft classification method. Afterward, a single hidden-layer feed-forward network with random input weights is adopted to learn the direct nonlinear mapping for the regression in each class. In the interpolation phase, the strategy of weighting multiple estimations is applied to enhance the interpolation performance. Furthermore, a refinement process is proposed to improve the interpolation performance with sharper edges and richer details. Extensive experimental results demonstrate that our proposed method is not only effective and efficient but also outperforms many state-of-the-art methods.

7 citations

Journal ArticleDOI
TL;DR: Experimental results show that the proposed method outperforms the state-of-the-art MMSE-directed linear interpolation schemes and works competitively with the state of the art learning-based ones.
Abstract: In this letter, we propose a minimum mean square error (MMSE) directed linear interpolation to compose the high-resolution image from a single low-resolution image. We build up our interpolation model by using some similar image patches selected according to the nonlocal geometric similarity. First, we use a two-stage search scheme to collect the matched patches inside the whole image. Second, a similarity scaling factor is used in the second search to refine the collected patches so as to help find a robust solution to the MMSE-directed interpolation. Third, our MMSE-directed interpolation is regularized by the involved reference patches to make the solved interpolation coefficients more reliable. Experimental results show that our proposed method outperforms the state-of-the-art MMSE-directed linear interpolation schemes and works competitively with the state-of-the-art learning-based ones.

7 citations


Cites background from "A new edge directed interpolation a..."

  • ...Since then, NEDI has been further improved via introducing various additional information [3]–[5]....

    [...]

Proceedings ArticleDOI
01 Oct 2017
TL;DR: This paper proposes a new expansion algorithm that preserves the edges of an object through modification of the neighborhood pixel values of the edges and generates higher contrast and less blurring and zigzag images with crisper appearances.
Abstract: In this paper, an improvement of the Canny edge-based image expansion algorithm is proposed. Our new expansion algorithm preserves the edges of an object. It generates the higher contrast and sharper images through modification of the neighborhood pixel values of the edges. In this method, we define two cases according to the orientations of the edges. In any diagonal orientation, we define two new operators to determine whether the diagonal direction is the left or right diagonal. For different cases, we propose different functions to process the neighborhood pixel values of the edges. Finally, we compare the expansion results and analyze the resulted image contrasts from different expansion algorithms. Our proposed expansion method generates higher contrast and less blurring and zigzag images with crisper appearances.

5 citations


Additional excerpts

  • ...Therefore, there are many algorithms [4-12] have been proposed to improve the subjective quality of the interpolated images by employing more accurate models [5]....

    [...]

Journal ArticleDOI
TL;DR: A novel image super‐resolution (SR) reconstruction method in the field of magnetic resonance imaging is proposed, which is based on a cross-modal edge‐preserving regularization integrating the internal gradient prior from the target‐modal image itself and the external gradient prior obtained by pre‐scan in many medical imaging scenes.
Abstract: In this article, we propose a novel image super‐resolution (SR) reconstruction method in the field of magnetic resonance imaging, which is based on a cross‐modal edge‐preserving regularization integrating the internal gradient prior from the target‐modal image itself and the external gradient prior from the reference‐modal image obtained by pre‐scan in many medical imaging scenes. The reference‐modal image is a high‐resolution guidance image that has much shareable information such as gradient orientation on edge regions, which can be used to improve the image resolution of the target modal. In addition, to be robust against the misalignment between the target‐modal image and reference‐modal image, a multimodal registration is incorporated in the SR reconstruction process. In this work, the proposed SR method can be formulated as an alternating optimization problem, that is, the target‐modal and reference‐modal images are alternately updated through iterations. Experimental results on simulated and realistic images show the superior performance of the proposed approach over several state‐of‐the‐art SR techniques.

4 citations


Cites methods from "A new edge directed interpolation a..."

  • ...But this method may cause some artifacts due to the incorrect estimation(9) or the mismatch of covariance.(11,12) Non-interpolation-based methods can be further grouped into two types: learning-based(13-16) and reconstruction-based methods....

    [...]

Journal ArticleDOI
TL;DR: A novel auto-regressive moving average (ARMA) model-based regularization term is introduced into the spare representation-based framework to deal with the single image super-resolution problem.

2 citations

References
More filters
Journal ArticleDOI
TL;DR: An overview of the technical features of H.264/AVC is provided, profiles and applications for the standard are described, and the history of the standardization process is outlined.
Abstract: H.264/AVC is newest video coding standard of the ITU-T Video Coding Experts Group and the ISO/IEC Moving Picture Experts Group. The main goals of the H.264/AVC standardization effort have been enhanced compression performance and provision of a "network-friendly" video representation addressing "conversational" (video telephony) and "nonconversational" (storage, broadcast, or streaming) applications. H.264/AVC has achieved a significant improvement in rate-distortion efficiency relative to existing standards. This article provides an overview of the technical features of H.264/AVC, describes profiles and applications for the standard, and outlines the history of the standardization process.

8,646 citations


"A new edge directed interpolation a..." refers methods in this paper

  • ...To give DCT coefficients the priority, the proposed method performs the quantization step using quantization table used MPEG [7]....

    [...]

  • ...It is widely used in image compression such as JPEG and MPEG....

    [...]

Journal ArticleDOI
TL;DR: A unified approach to the coder control of video coding standards such as MPEG-2, H.263, MPEG-4, and the draft video coding standard H.264/AVC (advanced video coding) is presented.
Abstract: A unified approach to the coder control of video coding standards such as MPEG-2, H.263, MPEG-4, and the draft video coding standard H.264/AVC (advanced video coding) is presented. The performance of the various standards is compared by means of PSNR and subjective testing results. The results indicate that H.264/AVC compliant encoders typically achieve essentially the same reproduction quality as encoders that are compliant with the previous standards while typically requiring 60% or less of the bit rate.

3,312 citations

Journal ArticleDOI
TL;DR: Simulation results demonstrate that the new interpolation algorithm substantially improves the subjective quality of the interpolated images over conventional linear interpolation.
Abstract: This paper proposes an edge-directed interpolation algorithm for natural images. The basic idea is to first estimate local covariance coefficients from a low-resolution image and then use these covariance estimates to adapt the interpolation at a higher resolution based on the geometric duality between the low-resolution covariance and the high-resolution covariance. The edge-directed property of covariance-based adaptation attributes to its capability of tuning the interpolation coefficients to match an arbitrarily oriented step edge. A hybrid approach of switching between bilinear interpolation and covariance-based adaptive interpolation is proposed to reduce the overall computational complexity. Two important applications of the new interpolation algorithm are studied: resolution enhancement of grayscale images and reconstruction of color images from CCD samples. Simulation results demonstrate that our new interpolation algorithm substantially improves the subjective quality of the interpolated images over conventional linear interpolation.

1,933 citations


"A new edge directed interpolation a..." refers methods in this paper

  • ...Compared to linear interpolation methods, the NEDI methods give outstanding interpolation images in aspects of sharpness and continuity of edge....

    [...]

  • ...The proposed method overcomes the existing problems of NEDI by considering multiple CB training windows....

    [...]

  • ...The NEDI method makes use of a fourth-order linear prediction to interpolate an unknown pixel from the four neighboring pixels (neighboring block NB), e.g. 1 1 2 1,2 1 2 2( ),2( ) 0 0 ˆ i j k l i k j l k l Y Ya+ + + + + = = =åå (1) The coefficients of the linear interpolation are the elements of the vector....

    [...]

  • ...RESULTS AND DISCUSSION The proposed algorithm has been compared with other interpolation algorithms in the literatures, including bilinear interpolation, the NEDI method....

    [...]

  • ...The NEDI algorithm is a two-step process that the first step estimates the unknown pixel Y2i+1,2j+1[ Fig1....

    [...]

Journal ArticleDOI
TL;DR: In this paper, the authors compared the performance of linear and cubic B-spline interpolation, linear interpolation and high-resolution cubic spline with edge enhancement with respect to the initial coordinate system.
Abstract: When resampling an image to a new set of coordinates (for example, when rotating an image), there is often a noticeable loss in image quality. To preserve image quality, the interpolating function used for the resampling should be an ideal low-pass filter. To determine which limited extent convolving functions would provide the best interpolation, five functions were compared: A) nearest neighbor, B) linear, C) cubic B-spline, D) high-resolution cubic spline with edge enhancement (a = -1), and E) high-resolution cubic spline (a = -0.5). The functions which extend over four picture elements (C, D, E) were shown to have a better frequency response than those which extend over one (A) or two (B) pixels. The nearest neighbor function shifted the image up to one-half a pixel. Linear and cubic B-spline interpolation tended to smooth the image. The best response was obtained with the high-resolution cubic spline functions. The location of the resampled points with respect to the initial coordinate system has a dramatic effect on the response of the sampled interpolating function?the data are exactly reproduced when the points are aligned, and the response has the most smoothing when the resampled points are equidistant from the original coordinate points. Thus, at the expense of some increase in computing time, image quality can be improved by resampled using the high-resolution cubic spline function as compared to the nearest neighbor, linear, or cubic B-spline functions.

844 citations


"A new edge directed interpolation a..." refers background in this paper

  • ...Therefore, many research studies tried to improve the visual quality of interpolated image at edge area using non-linear techniques [4]....

    [...]

Proceedings Article
01 Jan 2008
TL;DR: The implementation of the new algorithm (iNEDI, improved New Edge Directed Interpolation), even if computationally heavy (as the Li and Orchard’s method), obtained quality scores that are notably higher than those obtained with NEDI and other methods presented in the literature.
Abstract: In this paper we analyse the problem of general purpose image upscaling that preserves edge features and natural appearance and we present the results of subjective and objective evaluation of images interpolated using different algorithms. In particular, we consider the well-known NEDI (New Edge Directed Interpolation, Li and Orchard, 2001) method, showing that by modifying it in order to reduce numerical instability and making the region used to estimate the low resolution covariance adaptive, it is possible to obtain relevant improvements in the interpolation quality. The implementation of the new algorithm (iNEDI, improved New Edge Directed Interpolation), even if computationally heavy (as the Li and Orchard’s method), obtained, in both subjective and objective tests, quality scores that are notably higher than those obtained with NEDI and other methods presented in the literature.

123 citations


"A new edge directed interpolation a..." refers background in this paper

  • ...Varying the size of the training window such as [6] or increasing the number of neighboring pixels more than 4 such as [2, 3], the computational cost is too high....

    [...]