A survey on evaluation methods for image interpolation

Abstract

Image interpolation is applied to Euclidean, affine and projective transformations in numerous imaging applications. However, due to the unique characteristics and wide applications of image interpolation, a separate study of their evaluation methods is crucial. The paper studies different existing methods for the evaluation of image interpolation techniques. Furthermore, an evaluation method utilizing ground truth images for the comparisons is proposed. Two main classes of analysis are proposed as the basis for the assessments: performance evaluation and cost evaluation. The presented methods are briefly described, followed by comparative discussions. This survey provides information for the appropriate use of the existing evaluation methods and their improvement, assisting also in the designing of new evaluation methods and techniques.

Noise‐insensitive and edge‐preserving resolution upconversion scheme for digital image based on the spatial general autoregressive model

“…With the SGAR model, we interpolate n missing pixels y = (y 1 , y 2 , …, y n ) in an image window (see (4) at the bottom of next page)…”

Interpolation techniques versus F-transform in application to image reconstruction

“…RM SE = 0 represents identical images. 2 where Q op is the nearest known pixel with respect to the pixel at position (i, j). Because we want to minimize the complexity of this technique, we will ignore diagonal directions.…”

“…Their intensities are not presented in the original image and must be computed. The most common resampling technique is interpolation [2], [3], [4]. It is often used in, e.g., medical [1] image processing.…”

Improved image magnification algorithm based on Otsu thresholding

“…Many interpolation methods, including adaptive and non-adaptive methods, have been proposed in the last several decades. Classical non-adaptive methods, such as linear and cubic convolution methods [2] are preferable because of their computational simplicity, however, these methods are unable to adapt with varying local structures of a LR image, which causes undesirable artifacts such as blurring, blocking, and ringing around edges [3]. Several adaptive interpolation methods [2,[4][5][6][7][8][9][10][11][12][13][14][15], including edge-directed methods, have been proposed to address the problems of the aforementioned algorithms and to improve the perceptual quality of the interpolated images.…”