Remote-sensing crop disease identification method based on time phase and spectrum information and habitat condition

Abstract

The invention discloses a remote-sensing crop disease identification method based on time phase and spectrum information and a habitat condition. Multi-time-phase visible light-near infrared and thermal-infrared remote sensing images in a study area within a monitoring period are obtained and pretreatment is carried out on the images; with the remote sensing images after pretreatment, planting information of target crops is extracted by combining a certain kind of classification algorithm; and according to the visible light-near infrared and thermal-infrared remote sensing images, habitat information of disease monitoring is obtained, a disease area and a disease type of the target crops in the study area are determined by combining the time phase information of the target crops and the obtained habitat information, and severity of the disease is determined by using the spectrum information based on the determined disease area and disease type. According to the technical scheme, the remote sensing diagnosis precision of the crop disease can be improved effectively; and a reverse identification problem of remote sensing crop disease monitoring can be solved to a certain extent.

Description

technical field [0001] The invention relates to the field of remote sensing image processing and agricultural technology, in particular to a remote sensing recognition method for crop diseases based on time phase and spectral information and habitat conditions. Background technique [0002] Crop disease is an important biological disaster in agricultural production, and has become a dominant factor restricting agricultural high yield, high quality, high efficiency, ecology and safety. Restricted by production conditions and technological level, field-level sampling and field surveys have been the main means of disease identification and monitoring for a long time, and have played an important role in disease prediction and formulation of reasonable prevention strategies. However, this kind of method consumes a lot of manpower, material resources and time, and the sampling range and sample size are limited, so it is difficult to obtain a large-scale data at multiple points at...

AI Technical Summary

Problems solved by technology

When crops are stressed by diseases, drought and nutrition during the growth process, the spongy tissue will be destroyed, and the proportion of leaf pigments will change, which will cause the two chlorophyll absorption valleys in the visible light region to be inconspicuous, and the spectral reflection peak at 550nm will increase with leaf damage. When it becomes lower and flatter, the peak of the near-infrared light region is reduced or even disappears, and the wavy feature of the entire reflection spectrum curve is flattened. Therefore, by comparing the spectral curves of damaged and healthy crops, we can only determine whether the growth of the crops is affected. Coercion and its severity, as for the type of coercion, it is impossible to judge

[0004] The reverse remote sensing identification and diagnosis of crop diseases is very difficult due to the common spectral characteristics of different diseases, and the reverse remote sensing identification of crop diseases is the key to realizing large-scale aerospace remote sensing monitoring, and it is an unavoidable problem to monitor crop diseases using remote sensing images

Existing studies have shown that the occurrence of different diseases of the same crop is closely related to its growth cycle and habitat conditions. Temporal information and habitat factors are of great value for remote sensing monitoring of crop diseases. However, the comprehensive use of habitat, temporal and spectral information Disease identification has not been reported

Images