Crop planting mode recognition method

Abstract

The invention discloses a crop planting mode recognition method which comprises the following steps: inputting remote-sensing vegetation index time series data, and determining the vegetation growth threshold; optimizing parameters by selecting training samples to obtain the shortest length of the growing season of the crops, the longest length of the growing season of the crops and the smallest growth magnitude of the crops; extracting the vegetation growth information which is the number of the growing season of the crops in one year, the length of each growing season of the crops and the growth magnitude of the crops, and eliminating the non-crops; calculating the number of the growing season of the crops to obtain the multiple cropping index in one year; and finally determining the crop planting mode comprehensively according to the multiple cropping indexes of the crops in the last year, current year and next year. Due to the adoption of the technical scheme, the crop planting mode can be recognized directly according to the remote-sensing vegetation index time series data, and the problems the experience of calculating parameters is deficient, and the peak numbers do not correspond to the crop planting modes are solved. The crop planting mode recognition method has the advantages of strong noise resisting capability, less required parameters and strong versatility.

Description

Technical field [0001] The invention relates to the technical field of agricultural remote sensing, in particular to a method for identifying crop planting patterns. Background technique [0002] Cropping pattern (CP) is the spatial expression of crop rotation and is a generalization of the sequence of crops before and after crops. Crop planting mode is related to the full utilization of water, heat, light, soil fertility and other resources. It is an important technical link to improve field yield and total yield. It is of great significance for efficient and controllable agricultural management. With the world-wide redistribution of agricultural natural resources such as light, heat, and water, the continuous increase of the world's population, and the potential decline in the stock of arable land, planting patterns have also changed. Timely and accurate monitoring of agricultural planting patterns and their temporal and spatial changes is conducive to forecasting grain output...

AI Technical Summary

Problems solved by technology

[0016] The peak method has three disadvantages: first, the peak method is sensitive to noise and has high requirements for time series data filtering and denoising algorithms

Although the current filtering method can better eliminate some obvious noise, the filtered curve is not a completely smooth curve, and there are still some subtle noises, and the peak method will detect every peak

Second, the method depends on the researcher's experience and regional characteristics, and the method is not universally applicable

Third, there is not a one-to-one correspondence between the number of peaks and the planting mode, and it is inaccurate to represent the planting mode with the number of peaks

Therefore, the change in the number of peaks between two years is a "pseudo-change", and it is even more unreasonable to infer changes in the planting pattern.

[0017] In short, the existing methods for identifying crop planting patterns based on time-series data of remote sensing vegetation indices have low regional adaptability and weak universality.

Images