Corn-for-seed field remote sensing recognition method and device

Abstract

The invention provides a corn-for-seed field remote sensing recognition method and device, the method comprises following steps: S1, a multi-temporal first multispectral image in corn growth stage and a mono-temporal first panchromatic wave band image after removing male parent of the corn for seed in a target field are acquired, and the first multispectral image and cultivated land data are combined to generate a second multispectral image; S2, a vegetation index system of the second multispectral image is constructed, and a timing sequence image set is generated; S3, according to the timing sequence image set, a pre-constructed random forest classifier is used to recognize a first corn-for-seed field; and S4, textural features of the first corn-for-seed field are acquired from the first panchromatic wave band image, and a second corn-for-seed field is recognized from the first corn-for-seed field. According to the invention, recognition precision of corn-for-seed field is improved, overfitting is less, according to the textural feature difference between the corn-for-seed field and large-field corn, corn-for-seed field can be further recognized, and recognition precision of the corn-for-seed field can be further improved.

Description

technical field [0001] The invention relates to the field of remote sensing image processing, and more specifically, to a method and device for remote sensing identification of a seed production corn field. Background technique [0002] Hybrid seed production is an important link in the production of major crops such as corn and rice. Mastering the seed production area and yield of major crop seeds is the basis for balancing the seed supply and demand market. Crop seed production areas in my country are relatively concentrated, and county-level administrative regions are often used as the basic unit for monitoring and statistics of information such as area and yield. , growth trends, especially when disasters such as drought, heavy rain, and frost occur, it is of great significance to quickly and accurately obtain information such as the extent of crop damage and the affected area. Therefore, there is an urgent need for a method that can quickly and effectively obtain the dis...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the problem of poor recognition effect when using decision tree classifiers for crop remote sensing recognition in the above-mentioned prior art or at least partially solve the above problems, the present invention provides a method and device for remote sensing recognition of seed production cornfields

Images