Objective estimation of subjective image quality assessment using multi‐parameter prediction

Abstract

Objective evaluation of a subjective image quality assessment plays a significant role in the various image processing applications, such as compression, interpolation and noise reduction. The subjective image quality assessment does not only depend on some objective measurable artefacts, but also on image content and kind of distortions. Thus, a multi‐parameter prediction of the objective image quality assessment is proposed in this study. The prediction parameters are found minimising the mean square error related to the known subjective image quality measure (DMOS). This approach includes mostly used image quality metrics (peak signal‐to‐noise ratio, multi‐scale structural similarity image measure, feature similarity image measure, video quality measure) and two‐dimensional image quality metrics (2D IQM). The proposed multi‐parameter prediction has been verified on the test image database (LIVE) for compression, noise and blur distortions with available subjective image quality measures (DMOS). More reliable estimations are obtained using multi‐parameter prediction instead of only one measure. The best results are reached when an image content indicator is combined with the 2D IQM measure separately for different kinds of distortions.

Image Multi-feature Fusion with Restricted Boltzmann Machine and Neural Network

“…Restricted Boltzmann Machines, as an energy-based generative model, are widely used in data dimensionality reduction, data encoding and deep neural networks. The structure of a standard restricted Boltzmann machine is a symmetric bipartite graph consisting of binary-type visible layer nodes and hidden layer nodes, so it can only process binary input data [4][5]. To solve this drawback, Gaussian-Bernoulli Restricted Boltzmann Machines are designed to deal with real-type data (eg, image data).…”