Algorithmic Analysis of Edge Ranking and Profiling for MTF Determination of an Imaging System

TL;DR

This paper introduces a novel automated algorithm for edge ranking and detection in large image datasets to accurately determine the Modulation Transfer Function (MTF), enhancing imaging quality assessment especially for extensive satellite images.

Contribution

The paper proposes a new simple algorithm for edge profiling that automates MTF determination, suitable for large datasets and GPU implementation.

Findings

Effective edge detection and ranking in large datasets

Automated MTF computation improves imaging quality assessment

Algorithm optimized for GPU acceleration

Abstract

Edge detection is one of the most principal techniques for detecting discontinuities in the gray levels of image pixels. The Modulation Transfer Function (MTF) is one of the main criteria for assessing imaging quality and is a parameter frequently used for measuring the sharpness of an imaging system. In order to determine the MTF, it is essential to determine the best edge from the target image so that an edge profile can be developed and then the line spread function and hence the MTF, can be computed accordingly. For regular image sizes, the human visual system is adept enough to identify suitable edges from the image. But considering huge image datasets, such as those obtained from satellites, the image size may range in few gigabytes and in such a case, manual inspection of images for determination of the best suitable edge is not plausible and hence, edge profiling tasks have to be automated. This paper presents a novel, yet simple, algorithm for edge ranking and detection from image data-sets for MTF computation, which is ideal for automation on vectorised graphical processing units.