US 12,094,198 B2
Cloud platform-based garlic crop recognition method by coupling active and passive remote sensing images
Haifeng Tian, Kaifeng (CN); Yaochen Qin, Kaifeng (CN); Wei Shen, Kaifeng (CN); Boyan Zhou, Kaifeng (CN); and Yongjiu Wang, Kaifeng (CN)
Assigned to Henan University, Kaifeng (CN)
Filed by Henan University, Kaifeng (CN)
Filed on Nov. 22, 2021, as Appl. No. 17/531,826.
Application 17/531,826 is a continuation of application No. PCT/CN2021/097839, filed on Jun. 2, 2021.
Claims priority of application No. 202010995102.7 (CN), filed on Sep. 21, 2020.
Prior Publication US 2022/0092306 A1, Mar. 24, 2022
Int. Cl. G06V 20/10 (2022.01); G01S 13/90 (2006.01); G01S 19/14 (2010.01); G06F 18/21 (2023.01); G06F 18/243 (2023.01); G06F 18/25 (2023.01); G06N 5/02 (2023.01)
CPC G06V 20/188 (2022.01) [G01S 13/9004 (2019.05); G01S 13/9027 (2019.05); G01S 19/14 (2013.01); G06F 18/21 (2023.01); G06F 18/24323 (2023.01); G06F 18/256 (2023.01); G06N 5/02 (2013.01)] 9 Claims
OG exemplary drawing
 
1. A cloud platform-based garlic crop recognition method by coupling active and passive remote sensing images, comprising the following steps:
S1: retrieving moderate resolution imaging spectroradiometer (MODIS)-normalized difference vegetation index (NDVI) time series images of main garlic crop production areas in a target year on a Google Earth Engine cloud computing platform to obtain phenological information of garlic crops and other forest and grass vegetation according to the MODIS-NDVI time series images;
S2: retrieving Sentinel-2 time series images and Landsat-8 time series satellite images of the main garlic crop production areas in the target year on the Google Earth Engine cloud computing platform to obtain an optically synthetic image data set by combining the phenological information of the garlic crops;
S3: obtaining and recording geographic coordinate information of the garlic crops and winter wheat crops in the main garlic crop production areas by a hand-held global position system (GPS);
S4: constructing a decision tree model for optical image recognition of the garlic crops based on the optically synthetic image data set obtained in step S2 and the geographic coordinate information of the garlic crops obtained in step S3;
S5: classifying the optically synthetic image data set obtained in step S2 according to the decision tree model for the optical image recognition of the garlic crops obtained in step S4 to obtain an optical distribution diagram of the garlic crops;
S6: retrieving Sentinel-1 time series synthetic aperture radar satellite images of the main garlic crop production areas in the target year on the Google Earth Engine cloud computing platform to obtain radar image characteristics of the garlic crops and the winter wheat crops by combining the geographic coordinate information of the garlic crops and the winter wheat crops obtained in step S3;
S7: obtaining a radar synthetic image data set according to the radar image characteristics of the garlic crops and the winter wheat crops in step S6;
S8: constructing a decision tree model for radar image recognition of the garlic crops according to the radar synthetic image data set obtained in step S7 and the geographic coordinate information of the garlic crops obtained in step S3;
S9: classifying the radar synthetic image data set obtained in step S7 according to the decision tree model for the radar image recognition of the garlic crops obtained in step S8 to obtain a radar distribution diagram of the garlic crops; and
S10: coupling the radar distribution diagram of the garlic crops in step S9 with the optical distribution diagram of the garlic crops in step S5 on the Google Earth Engine cloud computing platform to obtain remote sensing-based recognition results of the garlic crops.