Landsat TM remote sensing image data cloud removal method and system

Abstract

The invention relates to a Landsat TM remote sensing image data cloud removal method. The method includes: inputting Landsat TM remote sensing image data; carrying out multi-scale segmentation on the above-mentioned Landsat TM remote sensing image data; establishing an operating feature ''ThickC'' for the above-mentioned Landsat TM remote sensing image data after the multi-scale segmentation according to a spectroscopic characteristic of thick clouds; classifying objects, which meet a threshold value condition, of the operating feature ''ThickC'' from "unclassified" into ''Thick Cloud''; establishing a relation feature ''ThinC'' for the above-mentioned Landsat TM remote sensing image data after the multi-scale segmentation according to a spectroscopic characteristic and a distribution characteristic of thin clouds; in remaining ''unclassified'' objects, utilizing a double threshold value classification method to classify the objects, which meet a condition, from the ''unclassified'' into ''Thin Cloud''; and uniformity classifying the above-mentioned objects, which are classified into the ''Thick Cloud'' and the ''Thin Cloud'', into ''Cloud'', and completing data cloud removal. The invention also relates to a Landsat TM remote sensing image data cloud removal system. According to the method and the system, influences of the clouds can be effectively reduced or removed, the amount of error classification and leaking classification is reduced, and the classification efficiency and precision are improved.

Description

technical field [0001] The invention relates to a method and system for removing clouds from Landsat TM remote sensing image data. Background technique [0002] Cloud occlusion is a very common phenomenon in the process of remote sensing data processing, especially when using remote sensing data for land cover monitoring and updating, due to cloud occlusion, the efficiency of data utilization is seriously affected. On the other hand, clouds and fog are the most numerous and most unstable targets in the atmosphere in South China. According to the retrieval office of the National Satellite Meteorological Center, due to the influence of clouds and fog, the average effective rate of NOAA / AVHRR remote sensing data in southern my country is less than 7%. At any moment when the sensor acquires data, it may encounter cloud and fog occlusion. Because of the existence of clouds, it takes a lot of energy for relevant workers to remove clouds when using satellite remote sensing images fo...

AI Technical Summary

Problems solved by technology

[0002] Cloud occlusion is a very common phenomenon in the process of remote sensing data processing, especially when using remote sensing data for land cover monitoring and updating, due to cloud occlusion, the efficiency of data utilization is seriously affected

On the other hand, clouds and fog are the most numerous and most unstable targets in the atmosphere in South China. According to the retrieval office of the National Satellite Meteorological Center, due to the influence of clouds and fog, the average effective rate of NOAA / AVHRR remote sensing data in southern my country is less than 7%. At any time when the sensor acquires data, it may encounter cloud and fog occlusion. Because of the existence of clouds, it takes a lot of energy for relevant workers to remove clouds when using satellite remote sensing images for drawing and other applications.

[0003] Due to climate reasons, it is difficult to obtain completely cloud-free remote sensing images. Most remote sensing images will be more or less affected by clouds when they are acquired. Therefore, removing the cloud has become a major problem faced by many Landsat TM application workers

[0004] Among the many cloud removal technologies, if there are large-scale thin clouds in the data, it is better to use the homomorphic filtering method. This is because the homomorphic filtering method combines frequency filtering and grayscale changes to separate clouds and background features. Finally, the influence of clouds is removed from the image, but this method involves the selection of filters and cut-off frequencies, sometimes some useful information is lost during the filtering process, and for the remote sensing data with a large amount of calculation, such as Landsat TM, the operation It is very inconvenient, and it is not possible to use this method for images with thick clouds

When there are large thick clouds in the processed data, the conventional method is to use the time averaging method, but this algorithm is only suitable for areas where the characteristics of the ground objects change little over time. For remote sensing images with medium temporal resolution such as Landsat TM It is often not possible to use this simple alternative algorithm

[0005] To sum up, in the case of limited spectral resolution, cloud coverage noise is difficult to remove by multispectral methods. For remote sensing platforms with low temporal resolution, cloud removal is an important factor that causes low utilization of remote sensing data. One, and seriously affect the subsequent use of remote sensing data, such as image recognition, change monitoring, supervised classification and other application problems

Images