An agricultural pest monitoring method and equipment based on hyperspectrum

Abstract

The embodiment of the invention discloses an agricultural pest monitoring method and equipment based on hyperspectrum. The problem that crop diseases and insect pests are difficult to accurately monitor in the prior art is solved. The method includes determining a position coordinate set of each small area to be measured; acquiring a pest and disease sensitive characteristic wave band corresponding to the to-be-detected area; acquiring a hyperspectral image of the to-be-detected crop area in the characteristic wave band, and marking latitude and longitude coordinates corresponding to the hyperspectral image; determining a to-be-measured small area corresponding to the hyperspectral image through the latitude and longitude coordinates; counting the number of damaged leaves in the to-be-detected small area, and when the number of the damaged leaves is greater than a first preset threshold value, inputting the hyperspectral image into a pest and disease damage type identification neural network model and a pest and disease damage grade estimation model to obtain the pest and disease damage type and pest and disease damage grade of the to-be-detected crop; and according to the pest type and the pest grade, monitoring the pest condition of the to-be-detected crop area. According to the method, the accuracy of crop disease and insect pest monitoring is improved.

Description

technical field [0001] The present application relates to the field of monitoring of agricultural diseases and insect pests, in particular to a method and equipment for monitoring agricultural diseases and insect pests based on hyperspectral. Background technique [0002] my country is a traditional large agricultural country, which has been dominated by farming for quite a long time, and is not a strong agricultural country in terms of agricultural technology. At present, my country's agricultural production is still dominated by the traditional production mode, and most of the agricultural management is based on experience, such as fertilization, irrigation, pest control, etc., resulting in a huge waste of human and financial resources. [0003] In particular, crop diseases and insect pests have the characteristics of many types, great impact, and frequent outbreaks. It is difficult to accurately monitor crops in the early stage of pests by relying on experience in agricu...

AI Technical Summary

Problems solved by technology

It is difficult to accurately monitor crops in the early stage of pests by relying on experience in agricultural management, which can easily lead to large-scale outbreaks of pests and diseases

In addition, the existing technology is inaccurate in monitoring the types of pests and diseases and the degree of infestation of pests and diseases. As a result, there are often problems that the types of pests and diseases do not match the types of medication, and the amount of pesticides is more or less, which seriously affects the operation effect of precision agriculture and is difficult to improve. agricultural output

Images