Method for identifying winter wheat

Abstract

The invention discloses a method for identifying winter wheat. On the basis of remote-sensing vegetation index time sequence data and phonological calendar data during a growth period of winter wheat, classification type setting, training sample selection, sample time sequence curve extraction, winter wheat identification feature selection and parametrization, optimal feature threshold determination and winter wheat identification model construction are carried out successively to extract spatial distrubtion information of the winter wheat in a range. The method is characterized in that spatial distribution information extraction of the winter wheat can be carried out based on the remote-sensing vegetation index time sequence data without any de-noising processing, so that the anti-time sequence data noise capability is high. The method has high stability and universality can be applied to remote-sensing extraction of distribution information of the winter wheat in a large range having a certain phonological difference.

Description

technical field [0001] The invention relates to the technical field of agricultural remote sensing, in particular to a remote sensing extraction method for spatial distribution information of winter wheat. Background technique [0002] Remote sensing image data has been widely used in the extraction of crop spatial distribution information due to its advantages of large coverage area and strong current situation. At present, crop identification methods based on remote sensing images can be divided into single-period image source methods and time-series image source methods. [0003] The single-period image source method is based on the single-period or few-period images of the best period, and uses the spectral difference between the crops to be identified and the background objects to realize the extraction of crop spatial distribution information. This type of method has high efficiency and strong operability, but there are also obvious deficiencies. On the one hand, suc...

AI Technical Summary

Problems solved by technology

However, affected by the noise of the time series data, the image features of the time series curves of certain crop pixels to be identified are prone to mutations at key nodes, and even the logical relationship with other image features on other nodes has also changed, causing the traditional method Crop recognition accuracy is low

Therefore, the traditional crop identification method based on time-series data often needs to rely on high-quality vegetation index time-series data to obtain high crop distribution information extraction accuracy, but due to the influence of atmospheric conditions (cloudy and rainy weather, etc.), we sometimes cannot obtain quality Generally high time-series data, which usually needs to be denoised before application

[0006] On the one hand, the denoising process can indeed remove some obvious noises and make the time series data smoother, but on the other hand, the denoising process also makes the crop phenological feature information reflected by the time series data be modified or used as noise information. Remove (for example, the bimodal feature of winter wheat is not obvious), and at the same time, new noise will be mixed in, thus affecting the selection of subsequent crop identification features

In addition, denoising processing also increases the complexity of crop identification and reduces work efficiency

Images