A decision tree classification method for middle rice information based on feature extraction from multi-temporal data

Abstract

According to the middle rice information decision tree classification method based on feature extraction in multi-temporal data in the present invention, the Gaofen-1 image data selected by it has the advantages of high spatial resolution and high temporal resolution. On this basis, the present invention It not only uses a variety of characteristic parameters used in the extraction of rice distribution from single-temporal images, but also combines the advantages of time-series analysis of multi-temporal images to organically combine multi-parameters and multi-temporal phases, and extract The distribution of middle rice. A variety of characteristic parameters can better eliminate non-target features. Multi-temporal analysis can help to eliminate misclassified features caused by foreign objects with the same spectrum, and can further extract target features. Decision tree classification is flexible, intuitive, and Features such as high efficiency. Therefore, combining these advantages can further improve the accuracy of mid-season rice extraction, which is of positive significance to both the national food security system and the commercial application of remote sensing in agriculture.

Description

technical field [0001] The invention relates to the field of surveying and mapping remote sensing, in particular to a middle rice information decision tree classification method based on feature extraction from multi-temporal data. Background technique [0002] Rice is one of the world's three major food crops, and its sown area accounts for 15% of the world's total arable land. my country is the world's largest rice producer, with a vast area of ​​rice cultivation, ranging from Hainan Province in the south, to Heilongjiang Province in the north, to Taiwan Province in the east, and to the Xinjiang Uygur Autonomous Region in the west. Different regions have different climates, and the varieties of rice planted are also different. According to the planting time, it can be divided into three categories: early rice, middle rice, and late rice. Different regions have different growth periods for early rice, middle rice, and late rice. Taking Hubei Province as an example, double...

AI Technical Summary

Problems solved by technology

However, the revisit cycle of TM image data is long, and it is difficult to obtain TM time-series images of rice in different periods, especially cloudy and rainy weather can easily have a negative impact on image quality

Images