Research On Intelligent Monitoring System Of Grassland Ecological Status Based On High Score Remote Sensing Images

High-score remote sensing technology has the advantages of long-term,high-resolution,large-width and high-efficiency data.In recent years,with the development of high-score remote sensing technology,a large number of fast and accurate data have been provided in various fields,such as grassland classification,grassland dynamic monitoring,planning of national major grassland ecological construction projects,grassland resources and ecological research and so on.For a long time,grassland ecosystem has been affected by both external and internal factors,resulting in grassland ecological degradation and weakening of grassland carrying capacity.The state proposes to establish a reward and compensation mechanism for grassland ecology and a series of policies such as judicial protection of grassland ecological environment have some problems such as omission in implementation and lack of supervision in the process of implementation.In view of the difficulties in the process of grassland ecological condition monitoring,such as high cost,heavy task and long cycle,an intelligent grassland ecological monitoring system based on high-score remote sensing images is studied and developed.the spatial distribution,temporal evolution and interaction of grassland surface cover were evaluated and analyzed,integrated with meteorological and human elements,and reflected the evolution and changes caused by climate change or human activities in different stages.And simulate and predict the evolution trend in the future.To a certain extent,reduce costs,improve efficiency and accuracy,effective analysis and decision-making.In this paper,the research data are 5-year high-score remote sensing images taken in the same area and at a fixed time.The main work is as follows:(1)Design and construct a hierarchical perception model.After data preprocessing,the VGG19 transfer learning algorithm is used to train the classification model,which is divided into five categories: edge,sand,grassland,woodland and road according to the actual situation of the study area,and then the confusion matrix is used to evaluate the classification effect of the classification model.(2)designing and constructing hierarchical perception model.The hierarchical image is processed and the feature database is established.Firstly,the regressionmodel of gray co-occurrence matrix and segmentation threshold is established,and the segmentation threshold is determined adaptively to segment the edge class,explain the logical relationship between land cover and detect the change,and visualize the evolution process of land cover.The K-Mean clustering segmentation algorithm is used to segment the road according to the road characteristics,and the remote sensing interpretation of the grassland road in the remote sensing image is analyzed.The Otsu method is used to segment the image of sandy land,calculate the desertification ratio of edge,road and desert image,and evaluate the degree of desertification according to the standard of desertification.(3)Design and construct the analysis and decision model.The ecological environment condition index is calculated and the future development trend is predicted by grayscale prediction model.Based on the data of statistical classification and hierarchical perception model,the dynamic changes of grassland surface cover type and area were analyzed automatically,and the overall evaluation of grassland ecological status was completed.Finally,the human-computer interface is designed,the above model is embedded in the system,the function of the above model is realized and tested,the effectiveness of the system is analyzed,and the overall evaluation of regional ecological status is completed.