Image Interpolation via Gradient Correlation-Based Edge Direction Estimation

Abstract

This paper introduces an image interpolation method that provides performance superior to that of the state-of-the-art algorithms. The simple linear method, if used for interpolation, provides interpolation at the cost of blurring, jagging, and other artifacts; however, applying complex methods provides better interpolation results, but sometimes they fail to preserve some specific edge patterns or results in oversmoothing of the edges due to postprocessing of the initial interpolation process. The proposed method uses a new gradient-based approach that makes an intelligent decision based on the edge direction using the edge map and gradient map of an image and interpolates unknown pixels in the predicted direction using known intensity pixels. The input image is subjected to the efficient hysteresis thresholding-based edge map calculation, followed by interpolation of low-resolution edge map to obtain a high-resolution edge map. Edge map interpolation is followed by classification of unknown pixels into obvious edges, uniform regions, and transitional edges using the decision support system. Coefficient-based interpolation that involves gradient coefficient and distance coefficient is applied to obvious edge pixels in the high-resolution image, whereas transitional edges in the neighborhood of an obvious edge are interpolated in the same direction to provide uniform interpolation. Simple line averaging is applied to pixels that are not detected as an edge to decrease the complexity of the proposed method. Applying line averaging to smooth pixels helps to control the complexity of the algorithm, whereas applying gradient-based interpolation preserves edges and hence results in better performance at reasonable complexity.

Stochastic Rounding for Image Interpolation and Scan Conversion

“…The fact that bilinear performed poorer than others in both (a) and (b) cases, confirms its inherent flaw of being the most interpolation blurriness productive. , where the original video frame rate was 1074, the bicubic interpolated scan conversion algorithm demonstrated strength that would normally be expected, as it normally produces better image quality than most non-adaptive interpolation algorithms [32], [33], [34], [35]. The mean scores were provided, in the legend, to quickly assess the performance of each interpolated scan conversion algorithm.…”

“…Now, considering (b), where the original video frame rate was 60, NNI-SR achieved higher PIQE scores than NNI-DR and in general, both NNI-DR and NNI-SR interpolated scan conversion algorithms achieved the best PIQE scores compared to the other two interpolation algorithms of the extra-pixel category. Considering (a), where the original video frame rate was 1074, the bicubic interpolated scan conversion algorithm demonstrated strength that would normally be expected, as it normally produces better image quality than most non-adaptive interpolation algorithms[32],[33],[34],[35]. The mean scores were provided, in the legend, to quickly assess the performance of each interpolated scan conversion algorithm.…”

An improvement of efficient nonlinear color image interpolator: Theory and FPGA implementation

“…These techniques often control interpolation by contour detection or image segmentation. An example of such techniques is given by Khan et al (2020). Other methods rely on fractal decompositions that allow the fine properties interpolated by the original sampling to be found (Kok and Tam, 2019).…”

Storage Optimization using Adaptive Thresholding Motion Detection

“…INTRODUCTION Image and video processing involves the transmission and storage of large data volumes in the form of frames and images. Several researchers [1][2][3] worked on reducing bandwidth requirements, whereas others [4] worked on optimizing storage utilization. In today's modern world, a huge amount of data is being generated [5], therefore, extracting the required information from large data volumes is a difficult task, while the storage of large data volumes is an issue [6].…”