A Fitting Method of Vegetation Parameters Based on Medium and High Resolution Remote Sensing

Abstract

The invention discloses a vegetation parameter fitting method based on middle-high resolution remote sensing. Due to the facts that used coarse resolution remote sensing data can be acquired easily, time resolution is very high, free shared data and products can be obtained, and the growth and development law difference of different vegetation types and the growth and development law difference in the vegetation of the same type are also used, when a study urges for the vegetation parameter in certain time, however, only the remote sensing data in another time can be acquired, through the method, the vegetation remote sensing parameter in the needed time can be simulated, and remote sensing study methods are enriched. For the reason that the middle-high resolution remote sensing is used for regional scale study, through the method, remote sensing data acquired at the study area in different time can be unified to the time which is needed by the study, and therefore the remote sensing data covering the entire study area in the same time can be acquired. Cloudless middle-high resolution remote sensing data in the needed time can reappear, and thus necessary data support can be provided for vegetation remote sensing correlation study work such as ecological remote sensing, environmental remote sensing and agricultural remote sensing.

Description

technical field [0001] The invention relates to a vegetation parameter fitting method based on medium and high-resolution remote sensing, which uses the time-space relationship of multi-scale remote sensing data to simulate unknown or missing data, such as NDVI and vegetation coverage. Background technique [0002] At present, coarse-resolution remote sensing images, such as MODIS, have high temporal resolution, and cloudless single-scene images or synthetic products can be obtained even in periods of poor weather conditions, while low spatial resolution leads to inconsistencies in research results. The accuracy is low, so it is mainly used in large-scale, such as province-wide research. In small areas, medium and high-resolution optical remote sensing develops rapidly and is widely used, but weather conditions such as cloud coverage are seriously affected, often causing missing images at the required time. For example, ETM+ data has an average of 35% cloud coverage in the w...

AI Technical Summary

Problems solved by technology

[0002] At present, coarse-resolution remote sensing images, such as MODIS, have high temporal resolution, and cloudless single-scene images or synthetic products can be obtained even in periods of poor weather conditions, while low spatial resolution leads to inconsistencies in research results. The accuracy is low, so it is mainly used in large-scale, such as province-wide research

In small areas, medium and high-resolution optical remote sensing develops rapidly and is widely used, but weather conditions such as cloud coverage are seriously affected, often causing missing images at the required time. For example, ETM+ data has an average of 35% cloud coverage in the world. Thus affecting the application of medium and high resolution remote sensing data

In practical applications, images at similar times or at the same time in previous years are often used instead. Due to differences in solar altitude angles and atmospheric conditions, there are errors in the research results

Aerial imagery can also be used instead, but it is expensive and difficult to afford for general research

There is also cloud removal processing such as filtering for cloud-covered remote sensing data, but it is mainly for images covered by thin clouds, and the effect is extremely poor when the cloud cover is thick

Images