Deep learning image evaluation method for video transmission quality

Abstract

The invention discloses a deep learning image evaluation method for video transmission quality. The method comprises the following steps: constructing a twin neural network for evaluating image quality to extract features, performing feature fusion on a target image block feature map and a reference image block feature map, and designing an image quality evaluation score calculation method and a loss function to update network parameters; setting training hyper-parameters of the twin neural network, wherein the training hyper-parameters comprise a learning rate, a learning attenuation rate, amaximum training step number, a learning rate attenuation step number, a batch training amount, a training image size, a single image sampling amount, data set (a training set, a verification set anda test set) distribution and the like; and designing an image region segmentation and block selection scheme of the target image and the reference image.

Description

technical field [0001] The invention relates to the field of image quality evaluation, in particular to an image quality evaluation method based on deep learning. Background technique [0002] Video images will have a great impact on the final transmission quality due to various reasons during the transmission process, so it is very important to correctly evaluate the image transmission quality. Many of the existing image quality evaluation technologies rely on the human visual system or natural image statistical computing models, or even artificial evaluation. The correlation between the evaluation score of the former and the average subjective opinion score of the human eye is sometimes relatively low, which is prone to inconsistent judgments; while the latter is very inefficient, prone to fatigue errors after long-term work, and affects the accuracy of judgment. Efficient, accurate and intelligent methods for evaluating the quality of video image transmission have import...

AI Technical Summary

Problems solved by technology

The correlation between the evaluation score of the former and the average subjective opinion score of the human eye is sometimes relatively low, which is prone to inconsistent judgments; while the latter is very inefficient, prone to fatigue errors after long-term work, and affects the accuracy of judgment. Efficient, accurate and intelligent methods for evaluating the quality of video image transmission have important practical significance

Images