Agricultural drought monitoring system based on remote sensing image and convolutional neural network

Abstract

The invention discloses an agricultural drought monitoring system based on a remote sensing image and a convolutional neural network. The system comprises a remote sensing image acquisition module, animage preprocessing module, an image storage module, an image feature extraction and classification module, a drought information analysis and processing module, a drought information display moduleand an information query and push module, which are are respectively responsible for high-resolution remote sensing image acquisition, remote sensing image preprocessing, remote sensing image storage,image feature extraction and classification, information analysis and processing, drought information visual display, and information query and pushing. According to the system, the convolutional neural network is adopted, the change problems of translation, rotation, scale zooming and other forms can be solved, and feature extraction is performed on the high-resolution satellite remote sensing image by using the powerful feature extraction capability of the convolutional neural network. The functions of drought information query, drought grading, drought information visual display, drought condition monitoring and prediction and the like can be achieved, and the requirement for agricultural drought monitoring under the current remote sensing image data growth background is met.

Description

technical field [0001] The invention belongs to the technical field of image processing and monitoring, in particular to an agricultural drought monitoring system based on remote sensing images and convolutional neural networks. Background technique [0002] Drought is a common global natural disaster due to the inconsistency between precipitation and local crop water demand. Nowadays, under the environment of increasing global warming, the uneven distribution of precipitation has further intensified, leading to a gradual increase in the frequency of droughts, and the problem of droughts needs to be resolved urgently. Due to the difficulty in detection at the early stage of its occurrence, the frequent occurrence, long duration and wide range of influence, it seriously affects the growth of crops and food production, and threatens social security. Therefore, it is very important to study drought monitoring methods suitable for agriculture and do a good job in drought monito...

AI Technical Summary

Problems solved by technology

[0003] (1) Due to the small number of observation sites, and the uneven distribution of sites and poor representation, and the need to consume a lot of manpower and material resources during monitoring

[0004] (2) Due to the use of index data that does not fully reflect the drought to determine the degree of drought in agricultural areas, the accuracy is low and the drought cannot be judged, resulting in the inability to effectively remedy the drought problem

[0005] (3) Due to the inability to effectively and accurately monitor large-scale crop planting areas and soil moisture, drought monitoring is insufficient and the accuracy and timeliness of monitoring results are not high

[0006] (4) Manually designed feature extraction methods, it is difficult to accurately describe specific application scenarios such as the data rules of agricultural areas, the generalization ability is poor, it is very difficult to select specific application models, and it is difficult to extract high-level abstract information of the target

[0007] In view of this, traditional agricultural drought monitoring methods have become increasingly difficult to meet the timeliness requirements for drought data processing in agricultural drought monitoring in my country

Images