Image noise reduction method and system

Abstract

The invention relates to an image noise reduction method and a system. The method comprises the steps of: converting an image to be processed into at least one image sub-layer by a client side, and transmitting the image sub-layers to a cloud computing system; performing noise reduction on the image sub-layers after the image sub-layers are received by the cloud computing system, and returning the image sub-layers subjected to noise reduction to the client side by the cloud computing system; reconstructing the image sub-layers subjected to noise reduction into an image after the image sub-layers subjected to noise reduction are received by the client side by the client side, thereby obtaining the image subjected to noise reduction. Since the image is layered, the complex image is decomposed into multiple image sub-layers for noise reduction and noise of each image sub-layer is reduced simultaneously via cloud computing, high-complexity computation can be performed to obtain high-quality images with noise reduced in the case of limited client side resources, and current demands for rapid and efficient image noise reduction can be satisfied in the case of increase of the image resolution and the number of images.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an image noise reduction method and system. Background technique [0002] In recent years, cloud computing has become one of the most concerned technologies in the field of IT (information technology), and it is also an important field that large enterprises and Internet IT construction are currently considering and investing in. The introduction of cloud computing has triggered new technological changes and new IT service models. Cloud computing is a comprehensive system that integrates software, application platforms and infrastructure through standardized IT technology and products, using the most cutting-edge technology and services. Users can quickly and conveniently request related services on demand through Internet technology. The main forms of cloud services can be summarized into three service models: SaaS (software as a service), PaaS (platform as a s...

AI Technical Summary

Problems solved by technology

The transform domain statistical modeling analysis method performs statistical modeling on the transform domain coefficients, which can achieve a better noise reduction effect, but requires more prior information to establish a suitable model for training, and the computational complexity is very high

Among the transform domain filtering methods, the shrinkage threshold denoising method based on wavelet transform is the most representative, but the shrinkage threshold in the transform domain is prone to distortion, which is called pseudo-Gibbs phenomenon

Images