A remotely sensed image denoising method based on incremental dictionary learning

A dictionary learning and remote sensing image technology, applied in the field of remote sensing image denoising based on incremental dictionary learning, to achieve the effect of suppressing noise

CN109214997AInactive Publication Date: 2019-01-15INST OF REMOTE SENSING & DIGITAL EARTH CHINESE ACADEMY OF SCI
5 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2019-01-15
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
Patent Text Reader

Abstract

The invention discloses a remotely sensed image denoising method based on incremental dictionary learning. The pixel intensity of pixel blocks extracted from a target image and pixel blocks extractedfrom a reference image has a linear relationship to utilize the similarity of edge features, and finally the coefficients are obtained as a restriction condition for solving a final objective function. Based on the local incremental dictionary learning, the noise in the remote sensing image is removed by using the similarity of the edge features, which can keep the characteristics of the target image while effectively suppressing the noise.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical field of remote sensing image processing, in particular to a remote sensing image denoising method based on incremental dictionary learning. Background technique

[0002] Many studies showed that in order to better remove the noise in remote sensing images, some reference information needs to be introduced. Compared with the method of denoising only using the information of the noisy remote sensing image itself, the denoising method of introducing reference information can use more information to denoise, so as to achieve a better denoising effect. The prior information can be introduced through the relationship between the pixel distribution of the target image and the pixel distribution of the reference image, and the prior information can also be introduced into the noise removal process in the wavelet domain. For multi-source remote sensing images, prior information can also be introduced in the gradient doma...

Examples

Embodiment Construction

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of the present invention belong to the protection scope of the present invention.

[0019] According to an embodiment of the present invention, a remote sensing image denoising method based on incremental dictionary learning is provided.

[0020] Such as figure 1 As shown, a remote sensing image denoising method based on incremental dictionary learning according to an embodiment of the present invention is characterized in that it includes the following steps:

[0021] Step 1: Extract similar pixel blocks from the target image and the reference image, and construct training...