US 12,405,259 B2
Rapid UAV-based monitoring and discrimination method for drought in summer maize based on chlorophyll content
Wenlong Song, Beijing (CN); Mengyi Li, Beijing (CN); Changjun Liu, Beijing (CN); Pu Zhou, Beijing (CN); Lang Yu, Beijing (CN); Yizhu Lu, Beijing (CN); Yun Liu, Beijing (CN); Wenjing Lu, Beijing (CN); Xiuhua Chen, Beijing (CN); and Long Chen, Beijing (CN)
Assigned to China Institute of Water Resources and Hydropower Research, Beijing (CN)
Filed by China Institute of Water Resources and Hydropower Research, Beijing (CN)
Filed on Dec. 23, 2024, as Appl. No. 18/991,999.
Application 18/991,999 is a continuation of application No. PCT/CN2023/119395, filed on Sep. 18, 2023.
Claims priority of application No. 202211140492.5 (CN), filed on Sep. 20, 2022.
Prior Publication US 2025/0130216 A1, Apr. 24, 2025
Int. Cl. G06Q 10/04 (2023.01); G01N 33/00 (2006.01); G06Q 50/02 (2024.01); G06T 7/00 (2017.01); G06T 7/90 (2017.01); G06V 20/10 (2022.01); G06V 20/17 (2022.01)
CPC G01N 33/0098 (2013.01) [G06V 20/17 (2022.01); G06V 20/188 (2022.01); G06V 20/194 (2022.01)] 4 Claims
OG exemplary drawing
 
1. A rapid unmanned aerial vehicle (UAV)-based monitoring and discrimination method for drought conditions in summer maize based on chlorophyll content, characterized by the following steps:
1) obtaining multispectral imagery data by an FL-81 quadcopter integrated with a multispectral camera and ground-measured chlorophyll content data by a chlorophyll content analyzer, calculating, by a processor, vegetation indices NDVI, RENDVI, and SAVI from the multispectral imagery data, wherein the multispectral camera is configured to capture multispectral aerial images, with a flight height set at 55 meters, corresponding to a ground resolution of 4 centimeters, and the camera is capable of capturing wavelengths in blue, green, red, red-edge, and near-infrared bands;
2) constructing, by the processor, Inversion Models for Chlorophyll Content at Different Drought Levels and Growth Stages of Summer Maize: Performing soil background pixel removal on the multispectral imagery data to extract pure vegetation index pixel values corresponding to the summer maize canopy; selecting the vegetation indices NDVI, RENDVI, and SAVI calculated in step 1) and constructing three types of regression equations (linear, exponential, and logarithmic) with the ground-measured chlorophyll content data at different growth stages; choosing the regression equation with the highest correlation with chlorophyll content for each growth stage as the optimal model equation for that stage; the different growth stages referred to are the jointing stage, heading stage, silking stage, and maturity stage of maize; specifically, the optimal model equation for the jointing stage is a logarithmic model regression equation between SAVI and chlorophyll content, the optimal model equation for the heading stage is a linear model regression equation between RENDVI and chlorophyll content, the optimal model equation for the silking stage is a logarithmic model regression equation between NDVI and chlorophyll content, and the optimal model equation for the maturity stage is an exponential model regression equation between RENDVI and chlorophyll content;
3) threshold determination for Chlorophyll Content at Different Drought Levels: Using the optimal model equations obtained in step 2) to invert the chlorophyll content at each growth stage and determine thresholds for chlorophyll content between different drought levels;
4) real-time Drought Level Discrimination: obtaining multispectral imagery of the test field through real-time monitoring and calculating the required vegetation indices; using the vegetation indices to invert the chlorophyll content at each growth stage by substituting them into the optimal model equations obtained in step 2); comparing the inverted chlorophyll content values with the thresholds determined in step 3) to assess the real-time drought level.