Large-area insect pest quantization monitoring system

Abstract

The invention provides a large-area insect pest quantization monitoring system. The large-area insect pest quantization monitoring system is characterized by comprising the following steps: acquiring a remote sensing image with a specific wave band by adopting aircraft-mounted or satellite-mounted hyperspectral remote sensing equipment, carrying out the necessary pre-treatment on the remote sensing image, eliminating the interference and measuring errors, then extracting the spectral signature in pixel-level size scale from the remote sensing image, and substituting the spectral signature into a multi-dimensional plant parameter model to invert a plant ecological index, so that an insect pest condition of a remote sensing detection zone is quantitatively determined according to the index.

Description

technical field [0001] The invention relates to agricultural information technology, more specifically, to a large-area insect pest quantitative monitoring system. Background technique [0002] The losses caused by pests to agricultural production are very huge. According to estimates, about 20% of the expected crop output is consumed or destroyed by pests every year. On the other hand, the pesticides used in large quantities to combat insect pests have brought about environmental pollution and food residue toxicity. Pests generally have a relatively long development period before the outbreak period. Therefore, through strict monitoring of the distribution, severity and change trend of pests, early prevention and intervention are not only effective means to limit the scope of pests and reduce crop losses. , can also reduce the amount of pesticide application as a whole. [0003] In the prior art, the technical means for monitoring pests mainly include trapping sampling me...

AI Technical Summary

Problems solved by technology

[0006] However, in the existing technology, there are still difficulties in the large-scale hyperspectral detection and analysis of pest-pest crops: the plant spectrum is sensitive to external environmental factors, and the image background, light, and meteorological conditions will have a non-negligible impact; the plant spectrum itself shows Sufficient time dynamics, but the lack of effective algorithms for data filtering and information extraction in the existing technology brings difficulties to the identification and quantitative research of pest disasters

Images