Gray mold early diagnosing system based on visable/near infrared multispectral imaging

Abstract

The invention discloses an early diagnosis system for plant leaves and canopy gray mold diseases based on visible / near-infrared multi-spectral imaging, which is characterized in that it includes a visible / near-infrared multi-spectral imager, an imager fixing device, an image acquisition card, a computer , Reflectivity dynamic calibration board and light source, visible / near-infrared multi-spectral imager are respectively connected to computer through RS-232 serial port line and image acquisition card data line, and image analysis and processing software is installed on the computer. The invention can realize fast, accurate, stable, real-time and non-destructive diagnosis of various plant gray mold diseases, and achieve early disease detection as much as possible; The amount of medicine caused by spraying all over, reduces the production cost and reduces the pollution.

Description

technical field [0001] The invention relates to an early diagnosis system of gray mold disease of plant leaves and canopy, in particular to an early diagnosis system of gray mold disease of plant leaves and canopy based on visible / near-infrared multi-spectral imaging. Background technique [0002] Plant gray mold is very harmful to plant growth. Diseases can not only lead to a decline in the yield and quality of economic plants, but also cause a large amount of investment in pesticides and other drugs and increase the cost of control, thereby increasing the cost of agricultural and forestry production and causing serious environmental pollution. Existing disease detection methods have the shortcomings of requiring a lot of manpower and material resources in sampling, measurement, and data analysis in actual agricultural and forestry production, and have poor real-time performance, which cannot meet the needs of specific production practices. There are interferences of subje...

AI Technical Summary

Problems solved by technology

Diseases can not only lead to a decline in the yield and quality of economic plants, but also cause a large amount of investment in pesticides and other drugs and increase in control costs, thereby increasing the cost of agricultural and forestry production and causing serious environmental pollution.

Existing disease detection methods have the shortcomings of requiring a lot of manpower and material resources in sampling, measurement, and data analysis in actual agricultural and forestry production, and have poor real-time performance, which cannot meet the needs of specific production practices.

There are interferences of subjective factors such as fatigue and emotion in naked eye observation; detection methods based on molecular level, such as bioassay technology, nucleic acid sequence analysis technology, molecular marker technology, PCR technology, etc., although relatively accurate, cannot be operated in real time and require professional Personnel operation is time-consuming and costly; the statistical prediction based on past disease occurrence data is not high enough in accuracy, lacks effectiveness, and is only suitable for rough estimation in a large geographical range, so it is difficult to predict or single plant

Images