Rosemary planting distribution high-resolution satellite remote sensing identification method

Abstract

The invention belongs to the technical field of high-resolution satellite remote sensing data ground object extraction and identification, and particularly relates to a rosemary planting distributionhigh-resolution satellite remote sensing identification method. The method comprises the following steps: 1) collecting multispectral data; 2) selecting a wave band set which is easy to distinguish inthe multispectral data; 3) carrying out atmospheric correction on the selected five wave bands, and carrying out histogram equalization processing; 4) constructing two vegetation index wavebands to increase a characteristic spectrum dimension of ground object pixels; 5) sequentially superposing and filtering seven wave bands contained in each image; 6) manually delineating a rosemary planting region sample, and calculating the average spectrum of all multi-dimensional vectors in the sample; (7) calculating a similarity probability of the standard spectrum and all pixel spectrums in a researchregion; (8) calculating the stability of a matching probability; and (9) making planting weight interpolation raster data and carrying out weight correction on the probability result. The method is low in labor cost, high in operation efficiency and extraction precision and good in recognition effect.

Description

technical field [0001] The invention belongs to the technical field of high-resolution satellite remote sensing data surface feature extraction and recognition, and in particular relates to a high-resolution satellite remote sensing recognition method for rosemary planting distribution using a multidimensional time-series spectral feature similarity matching algorithm. Background technique [0002] At present, the method of using high-resolution satellite remote sensing data to identify and classify surface crops has become a common method for agricultural planting structure investigation and information acquisition, especially for staple food crops with a wide distribution area, a large proportion of planting, and a simple planting structure ( Such as wheat, rice, etc.) have made great progress in identification and extraction technology. Compared with the identification and extraction of staple food crops, due to the characteristics of less planting, scattered distribution...

AI Technical Summary

Problems solved by technology

[0003] Especially for the extraction of rosemary medicinal crops, which are rarely planted, due to the sparse planting, irregular harvest period, and scattered planting by farmers in the area, the pixel mixing degree and spectral interference are serious. Conventional high-resolution High rate satellite remote sensing data cannot be effectively identified. Therefore, there are generally few applied researches on the extraction of rosemary based on commonly used satellite remote sensing data. The main method is to use drone flight remote sensing data to extract and monitor concentrated planting cultivation , the distribution of rosemary planted in a wide range is mainly based on regional aerial flight identification and artificial aerial photo delineation. The project progress is slow and the operation cost is high. It is difficult to take advantage of the advantages of satellite remote sensing identification technology.

Images