Multi-time sequence index change trend-based automatic grain for green recognition method

Abstract

The invention relates to a multi-time sequence index change trend-based automatic grain for green recognition method. The method comprises the following steps of: establishing a multi-year daily enhanced vegetation index time sequence data set of a research area, and extracting 5 time sequence indexes such as abundance, time sequence dispersion, growing period length, high-quantile sustainabilityand low-quantile sustainability pixel by pixel and year by year on the basis of enhanced vegetation index time sequence data; detecting multi-year change trends of the five time sequence indexes in sequence; and establishing an automatic grain for green recognition flow chart on the basis of the change trends of the five time sequence indexes, so as to realize automatic grain for green recognition. According to the method, vegetation growth states are divided according to quantile values, a plurality of time sequence indexes are designed from the aspects of coverage time, average state and change amplitude to describe the growth features of single-season crops, multi-season crops and forest vegetation, so that changes from crops to forests are indicated by utilizing the change trends of multi-time sequence indexes and then grain for green areas are automatically recognized; and the method has good flexibility and extensibility.

Description

technical field [0001] The invention relates to the field of remote sensing information processing, in particular to an automatic identification method for returning farmland to forests based on the changing trend of multiple time series indicators. Background technique [0002] All things are born in the soil. Cultivated land is the most basic natural resource for human survival and development. With the growth of the population, the demand for food from the land has intensified. Due to the scarcity of arable land in mountainous and hilly areas, the blind reclamation of planting slopes has brought serious water and soil erosion problems. Cultivation in sandy land in arid areas has caused frequent natural disasters such as drought and sandstorms, seriously threatening ecological security. Therefore, by investing a large amount of funds, the policy of returning farmland to forests will be implemented in a planned and step-by-step manner. [0003] According to reports, sin...

AI Technical Summary

Problems solved by technology

However, due to the interference of "intra-class heterogeneity and inter-class similarity" of different land cover spectra, as well as the influence of many factors such as the experience of remote sensing classifiers, classifiers, the richness of related data, and the characteristics of the study area, land Overlay automatic mapping accuracy is difficult to guarantee

[0004] The traditional land cover change detection is mostly based on the method of change detection after classification. The disadvantage of this method is that: multiple classifications bring cumulative errors, and the area where land cover change occurs is relatively small, which is easy to cause misclassification or omission.

Images