Forest dynamic change mode automatic extraction method

Abstract

The invention relates to a forest dynamic change mode automatic extraction method. Time series data sets of a vegetation index enhanced day-by-day for many years and a snow accumulation index of a research area are established. Each pixel extracts indexes such as vegetation cover strength, dispersion, sustainability and the like and extracts a brightness index year by year based on the enhanced vegetation index time series data. Each pixel detects secular variation trends of the above indexes in sequence and establishes a forest dynamic change mode recognition flow chart based on the variation trends of the indexes such as the vegetation cover strength, the dispersion, the sustainability, the brightness and the like, and finally, the purpose of forest dynamic change automatic monitoring is achieved. According to the invention, a plurality of indexes are designed from aspects such as the cover time, the average state, the variation amplitude and the like, based on the secular variation trends of the indexes, the forest dynamic change mode is recognized effectively, training data of a known sampling area and man-machine interaction are not needed, the robustness is good, the classification precision is high, and the automation extent and the anti-interference capability are high.

Description

technical field [0001] The invention relates to the technical field of remote sensing classification, in particular to a method for automatically extracting forest dynamic change patterns. Background technique [0002] As the lungs of the earth, forests continuously deliver the oxygen we need and a large amount of organic matter. The survival and development of human beings are closely related to the forest ecosystem. However, since modern times, due to the uncontrolled deforestation of human beings, the forests have been greatly reduced. my country is a forest-poor country, and the per capita forest area is less than 1 / 4 of the world's. During the period of rapid economic development in my country in the 21st century, the demand for timber will continue to grow. Therefore, efficient and rapid monitoring of dynamic changes in forests is crucial for the rational use and protection of natural resources. Traditional forest survey methods are difficult to meet the needs of cu...

AI Technical Summary

Problems solved by technology

Although the existing related methods show the characteristic advantages of such methods, there are still some shortcomings in the following aspects: (1) the algorithm is relatively complex, such as usually involving the decomposition and reconstruction of time series; (2) the algorithm The degree of automation needs to be improved; (3) Most of them focus on the increase or decrease of forests, and cannot effectively identify the dynamic change patterns of forests

Images