Method for measuring maximum signal to noise ratio of remote sensing image

Abstract

The invention provides a method for measuring the maximum signal to noise ratio of a remote sensing image. The method for measuring the maximum signal to noise ratio of the remote sensing image comprises the steps that an image sample area is extracted automatically through a sliding window, traversal analysis is conducted on an image to be tested, multiple sample statistic extreme values of the image are obtained, and then the maximum signal to noise ratio of the image to be tested is calculated by combining the sample statistic extreme values. According to the method for measuring the maximum signal to noise ratio of the remote sensing image, the detail information of the image can be effectively mined, the defect that according to existing main methods, the requirements for a sample area for measurement of the maximum signal to noise ratio are high is overcome, and the universality is achieved; in addition, measurement is conducted automatically, the influence of human factors is avoided, the unique maximum signal to noise ratio measurement result is obtained, and the repeatability is high. Based on all the advantages, the method has the broad application prospect and high application value in extraction and evaluation of information of remote sensing images.

Description

technical field [0001] The invention belongs to the field of image measurement and relates to a method for measuring the maximum signal-to-noise ratio of remote sensing images. Background technique [0002] Signal-to-noise ratio (SNR) is a key indicator of remote sensing image quality evaluation. The current mainstream remote sensing image signal-to-noise ratio measurement method, such as the reference "GJB5088-2002 resource satellite on-orbit image quality evaluation method"; "QJ20099.1- 2012 Land Observation Satellite Remote Sensing Image Quality Evaluation Method"; "Gaofen-1 satellite image quality on-orbit test evaluation" mentioned that it is necessary to first artificially select a uniform sample area (block), and then press the row / column / block of the image to be tested Calculate the mean and standard deviation, the ratio of the two is the SNR of the row / column / block, and the mean or block SNR of all rows / columns in the sample area is the image SNR. Under the conditi...

AI Technical Summary

Problems solved by technology

Under the condition of a high-brightness uniform sample area, this type of method can measure the maximum signal-to-noise ratio of the image well. This kind of signal-to-noise ratio measurement method is not universal, and the signal-to-noise ratio of most images, especially cities and surrounding areas that are highly affected by human activities, which is the most utilized remote sensing image, cannot Measurement

Even if there is a high-quality sample area on the image, the selection of the sample area is heavily dependent on manual intervention. For the same image, the measurement results are affected by the user's human factors, and the repeatability is not good.

Some researchers proposed an automatic measurement algorithm in the article "Image Signal-to-Noise Ratio Algorithm and Its Application in CBERS-1 Image Evaluation", but the actual effect is not ideal, and it has not been transformed into a practical technology for popularization and application.

Images