Method and system for evergreen forest land identification based on remote sensing information

Abstract

The embodiment of the present invention provides an evergreen woodland identification method and system based on remote sensing information, which makes full use of the first spatial resolution remote sensing image and the first resolution multi-view remote sensing image on the time series, and introduces evergreen vegetation and The concept of very green vegetation discrimination index is used as a standard to determine the proportion of evergreen forest land in the area corresponding to each pixel position, so that the final area of ​​evergreen forest land in the target area has a higher accuracy.

Description

technical field [0001] The invention relates to the technical fields of ecology and remote sensing, and more specifically, to a method and system for identifying evergreen woodland based on remote sensing information. Background technique [0002] For urban vegetation ecological environment management, carry out accurate evergreen forest land information extraction on complex underlying surfaces, which has great significance for urban garden management, freezing disaster risk prediction, oxygen release and carbon sequestration, nutrient maintenance, biodiversity protection and other forest ecological benefit estimation. Significance. However, the underlying surface of urbanized areas is relatively complex, including bridges, residential, office and other construction areas, vegetation coverage areas such as grassland, deciduous forest land and evergreen forest land, wet areas such as rivers and lakes, cement roads, dirt roads and asphalt roads It is difficult to obtain ever...

AI Technical Summary

Problems solved by technology

However, urban green space includes not only evergreen forest land, deciduous forest land, but also grassland, etc. Due to the complexity of the urban underlying surface, it is difficult to find regularity in the distribution of forest land and grassland, and the vigorous growth period of vegetation at medium spatial resolution is different. The spectral information of forest land and grassland is difficult to effectively distinguish between forest land and grassland, and thus it is impossible to accurately obtain evergreen forest land information

[0004] In addition, it is also difficult to extract evergreen forest land information with medium spatial resolution remote sensing technology. The main reason is that during the vigorous growth period, the spectra of evergreen forest land and deciduous forest land are similar. It is difficult to distinguish evergreen forest land from deciduous forest land from the perspective of spectrum, resulting in greater uncertainty in the extraction accuracy of evergreen forest land information; the extraction of evergreen forest land information by using time-series remote sensing images, due to the phenology of forest land Due to differences and limitations of existing models, it is difficult to obtain accurate distribution information of evergreen forest land

For remote sensing data with high spatial resolution, due to the phenological differences of different types of forest land, it is not feasible to use high spatial resolution time series analysis method to extract large-scale evergreen forest land information in terms of cost, workload and implementation effect. best choice

In terms of applied model methods, it mainly includes land use classification method, remote sensing index method, etc. Due to the large computational complexity of land use classification methods and the similarity between the spectra of evergreen forest land and deciduous forest land, it is difficult to obtain satisfactory results. effect; while remote sensing index methods, such as vegetation index, are difficult to effectively identify

Images