On Evaluation Criterion of Image Super-Resolution Reconstruction

The image super-resolution and its reconstruction techniques are introduced, the super-resolution evaluation criterion is preliminarily discussed, the minimizing norm of image difference gradient modular square sum, frequency domain method and profile density curve method are proposed and simulation tests and analysis are carried out in this paper. It is shown from the tests that these methods are superior to the traditional methods of objective evaluation and can be better in accordance with the subjective evaluation results.