Crop Disease Source Location And Monitoring System Based On Diffractive Light Identification Gas Sensor Network

In 2022,crop diseases and insect pests are on the reoccurrence trend with an estimated area of 2.01 billion.It is important to realize the early detection of crop diseases.Hyperspectral images and satellite remote sensing are widely used to detect crop diseases at home and abroad.These methods are mainly based on changes in crop phenology after disease outbreaks,and the above methods are lagging behind.Crop fungal diseases are transmitted by disease spores in the air.If the information of disease spores can be obtained from transmission routes,the early warning of crop diseases and the location prediction of disease sources can be realized.The main problems in obtaining disease spore transmission information include:(1)In the early stage of crop fungal diseases,the concentration of disease spores in the air is low.Microscopic imaging is limited by the traditional optical laws and the field of view is small,which is difficult to meet the needs of low-concentration spore detection.(2)The location of the crop disease outbreak source is random.The disease spores are susceptible to crop shading and airflow during transmission,which is difficult to effectively estimate the specific location of the disease source.Therefore,a crop disease source location and monitoring system based on diffractive light identification gas sensor network is proposed in this paper.The work was carried out in the following aspects:(1)Based on scalar diffraction theory,the mechanism of diffraction hologram image formation was explored.A large field-of-view diffraction imaging structure is designed and coupled with the enrichment structure to realize disease spore enrichment and image collection.(2)The diffraction image was preprocessed by light compensation and denoising.The appropriate algorithm was selected to reconstruct the image.The contour of the sample was extracted by threshold segmentation and morphology.The samples were counted by feature detection.(3)Based on the propagation theory of airborne spores,the simulation software was used to study the disease spore propagation law in undirected wind and directed wind.The sentinel search algorithm of disease source was proposed to locate disease source.(4)A sensor node integrating diffraction image and environmental information collection and transmission was designed and manufactured.The Internet of Things platform was built for data flow.The computer client was written to receive the data of the node.The image detection algorithm and the disease source location algorithm were invoked to realize the sample count and the prediction of the disease source location.(5)The polystyrene microspheres were used as samples to verify the feasibility of node enrichment and diffraction function,the accuracy of image detection algorithm,the propagation law of disease source,the accuracy of positioning algorithm,and the communication stability of the positioning monitoring system by experiments.The experimental results show that the designed node can realize the enrichment and diffraction image collection functions of the samples.The image detection algorithm is used to automatically identify the sample.The accuracy is more than 85%compared with the manual counting result.The effective field area of diffraction imaging is 40.4 times higher than conventional microscope under 40 x objective lens.Compared with the experimental data,the propagation rule of disease spores obtained by simulation shows a linear correlation with high accuracy.The sentinel search algorithm is used to locate disease sources on 2m×2m experimental site.The average relative errors of the initial localization of the disease source are 12.86% and 24.32%in undirected wind.The average relative errors of the secondary localization of the disease source are 5.95% and 10.61% in undirected wind.By testing the bit error rate and the time delay of the positioning monitoring system,the communication performance of the system is good,which is able to meet the actual needs.In conclusion,the crop disease source location and monitoring system based on diffractive light identification gas sensor network can realize the early detection of disease and the prediction of disease source location.