Granted Patent B1

US 12,260,636 · App. 18/286,235 · Granted Mar 25, 2025

Inventors: Wengang Zheng (Beijing, CN); Zhonglili Zhang (Beijing, CN); Chunjiang Zhao (Beijing, CN); Liping Chen (Beijing, CN); Shirui Zhang (Beijing, CN); Xin Zhang (Beijing, CN); Xinyue Lv (Beijing, CN)

Assignee: INTELLIGENT EQUIPMENT RESEARCH CENTER, BEIJING ACADEMY OF AGRICULTURE AND FORESTRY SCIENCES

G06V20/188G01N33/0098G06T7/0012G06T7/11G06T7/174G06V10/765G06V20/68G06V20/70G06T2207/10048G06T2207/30004G06T2207/30188

Abstract

Claims (81)

1. A decision-making method for water and fertilizer stress of crops, comprising:

determining a current crop coefficient and canopy temperature of a crop to be identified on the basis of a first crop image and a second crop image;

determining a water stress state and a nutrient stress state on the basis of the current crop coefficient and canopy temperature of the crop to be identified; and

processing the water stress state and the nutrient stress state with a fuzzy control algorithm, and determining a water and fertilizer stress decision result of the crop to be identified, wherein

the decision result comprises an irrigation amount and/or a topdressing amount for the crop to be identified in a next cycle; the first crop image is a visible light image obtained by photographing the crop to be identified in full view with a terminal; the second crop image is a thermal infrared image obtained by photographing the crop to be identified in full view with an infrared module connected to the terminal;

the processing the water stress state and the nutrient stress state with a fuzzy control algorithm, and determining a water and fertilizer stress decision result of the crop to be identified comprises:

converting the water stress state and the nutrient stress state into linguistic variables, inputting the linguistic variables into a fuzzy controller, and obtaining an irrigation instruction output from the fuzzy controller;

setting, in response to determining that first identifier information and second identifier information that are carried by the irrigation instruction are non-null and null respectively, an irrigation amount determined according to the crop coefficient and a reference crop evapotranspiration amount as the water and fertilizer stress decision result;

setting, in response to determining that first identifier information and second identifier information that are carried by the irrigation instruction are null and non-null respectively, a topdressing amount determined according to the nutrient stress state and species information of the crop to be identified as the water and fertilizer stress decision result; and

setting, in response to determining that neither of first identifier information and second identifier information that are carried by the irrigation instruction is null, an irrigation amount determined according to the crop coefficient and a reference crop evapotranspiration amount and a topdressing amount determined according to the nutrient stress state and species information of the crop to be identified as the water and fertilizer stress decision result; and

the fuzzy controller is created on the basis of a corresponding fuzzy rule between an input fuzzy set and an output fuzzy set; the input fuzzy set is determined according to a membership function set on the basis of fuzzy scales obtained by conducting fuzzy division on the water stress state and the nutrient stress state; the output fuzzy set is determined according to a membership function set on the basis of strategy scales of a plurality of water and fertilizer stress states obtained by clarifying a fuzzy language; the fuzzy rule is created by determining corresponding water and fertilizer stress states according to different water stress states and nutrient stress states; identifier information carried by the irrigation instruction is determined according to the water and fertilizer stress state; the first identifier information is configured to indicate whether to conduct irrigation; and the second identifier information is configured to indicate whether to conduct topdressing.

2. The decision-making method for water and fertilizer stress of crops according to claim 1 , wherein the determining a current crop coefficient and canopy temperature of a crop to be identified on the basis of a first crop image and a second crop image comprises:

segmenting the first crop image with an excess green algorithm, and obtaining a crop canopy image;

determining the crop coefficient according to a canopy coverage ratio analyzed from the crop canopy image; and

determining the canopy temperature on the basis of the crop canopy image and the second crop image.

3. The decision-making method for water and fertilizer stress of crops according to claim 2 , wherein the determining the canopy temperature on the basis of the crop canopy image and the second crop image comprises:

extracting outline coordinates of the crop canopy image and the second crop image separately with a binary image connected component algorithm, and obtaining a first feature point set and a second feature point set; and

obtaining an infrared canopy image by matching the first feature point set and the second feature point set, and then extracting the canopy temperature from the infrared canopy image.

4. The decision-making method for water and fertilizer stress of crops according to claim 2 , wherein the determining the crop coefficient according to a canopy coverage ratio analyzed from the crop canopy image comprises:

inputting the crop canopy image into an analytical model, and obtaining the crop coefficient output from the analytical model, wherein

the analytical model is trained according to a sample crop canopy image and a canopy coverage ratio label and a crop coefficient label that are set in the sample crop canopy image; the analytical model comprises a ratio conversion layer and a ratio analysis layer;

the ratio conversion layer is configured to determine the canopy coverage ratio on the basis of the crop canopy image; and

the ratio analysis layer is configured to estimate the crop coefficient on the basis of the canopy coverage ratio.

5. The decision-making method for water and fertilizer stress of crops according to claim 1 , wherein the determining a water stress state and a nutrient stress state on the basis of the crop coefficient and canopy temperature of the crop to be identified comprises:

setting a crop water stress index determined according to the canopy temperature as the water stress state; and

setting a normalized difference vegetation index determined according to the crop coefficient as the nutrient stress state.

6. The decision-making method for water and fertilizer stress of crops according to claim 1 , wherein after the processing the water stress state and the nutrient stress state with a fuzzy control algorithm and determining a water and fertilizer stress decision result of the crop to be identified, the method further comprises:

controlling, in response to determining that the water and fertilizer stress decision result only comprises the irrigation amount, an irrigation apparatus to add water of the irrigation amount to the crop to be identified in the next cycle according to a first control instruction;

controlling, in response to determining that the water and fertilizer stress decision result only comprises the topdressing amount, a fertilization apparatus to add fertilizers of the topdressing amount to the crop to be identified in the next cycle according to a second control instruction; and

controlling, in response to determining that the water and fertilizer stress decision result only comprises the topdressing amount, an irrigation apparatus to add water of the irrigation amount to the crop to be identified in the next cycle according to a third control instruction, and a fertilization apparatus to add fertilizers of the topdressing amount to the crop to be identified in the next cycle.

7. A mobile phone terminal, comprising a mobile phone terminal body and an infrared module that are in communication connection with each other, wherein the mobile phone terminal body comprises a memory, a processor, and a computer program stored in the memory and runnable in the processor, and the processor implements the decision-making method for water and fertilizer stress of crops according to claim 1 when executing the computer program.

8. The mobile phone terminal according to claim 7 , wherein the determining a current crop coefficient and canopy temperature of a crop to be identified on the basis of a first crop image and a second crop image comprises:

segmenting the first crop image with an excess green algorithm, and obtaining a crop canopy image;

determining the crop coefficient according to a canopy coverage ratio analyzed from the crop canopy image; and

determining the canopy temperature on the basis of the crop canopy image and the second crop image.

9. The mobile phone terminal according to claim 8 , wherein the determining the canopy temperature on the basis of the crop canopy image and the second crop image comprises:

extracting outline coordinates of the crop canopy image and the second crop image separately with a binary image connected component algorithm, and obtaining a first feature point set and a second feature point set; and

obtaining an infrared canopy image by matching the first feature point set and the second feature point set, and then extracting the canopy temperature from the infrared canopy image.

10. The mobile phone terminal according to claim 8 , wherein the determining the crop coefficient according to a canopy coverage ratio analyzed from the crop canopy image comprises:

inputting the crop canopy image into an analytical model, and obtaining the crop coefficient output from the analytical model, wherein

the analytical model is trained according to a sample crop canopy image and a canopy coverage ratio label and a crop coefficient label that are set in the sample crop canopy image; the analytical model comprises a ratio conversion layer and a ratio analysis layer;

the ratio conversion layer is configured to determine the canopy coverage ratio on the basis of the crop canopy image; and

the ratio analysis layer is configured to estimate the crop coefficient on the basis of the canopy coverage ratio.

11. The mobile phone terminal according to claim 7 , wherein the determining a water stress state and a nutrient stress state on the basis of the crop coefficient and canopy temperature of the crop to be identified comprises:

setting a crop water stress index determined according to the canopy temperature as the water stress state; and

setting a normalized difference vegetation index determined according to the crop coefficient as the nutrient stress state.

12. The mobile phone terminal according to claim 7 , wherein after the processing the water stress state and the nutrient stress state with a fuzzy control algorithm and determining a water and fertilizer stress decision result of the crop to be identified, the method further comprises:

controlling, in response to determining that the water and fertilizer stress decision result only comprises the irrigation amount, an irrigation apparatus to add water of the irrigation amount to the crop to be identified in the next cycle according to a first control instruction;

controlling, in response to determining that the water and fertilizer stress decision result only comprises the topdressing amount, a fertilization apparatus to add fertilizers of the topdressing amount to the crop to be identified in the next cycle according to a second control instruction; and

controlling, in response to determining that the water and fertilizer stress decision result only comprises the topdressing amount, an irrigation apparatus to add water of the irrigation amount to the crop to be identified in the next cycle according to a third control instruction, and a fertilization apparatus to add fertilizers of the topdressing amount to the crop to be identified in the next cycle.

13. A non-transient computer-readable storage medium, storing a computer program, wherein the computer program implements the decision-making method for water and fertilizer stress of crops according to claim 1 when being executed by a processor.

14. The non-transient computer-readable storage medium according to claim 13 , wherein the determining a current crop coefficient and canopy temperature of a crop to be identified on the basis of a first crop image and a second crop image comprises:

segmenting the first crop image with an excess green algorithm, and obtaining a crop canopy image;

determining the crop coefficient according to a canopy coverage ratio analyzed from the crop canopy image; and

determining the canopy temperature on the basis of the crop canopy image and the second crop image.

15. The non-transient computer-readable storage medium according to claim 14 , wherein the determining the canopy temperature on the basis of the crop canopy image and the second crop image comprises:

extracting outline coordinates of the crop canopy image and the second crop image separately with a binary image connected component algorithm, and obtaining a first feature point set and a second feature point set; and

obtaining an infrared canopy image by matching the first feature point set and the second feature point set, and then extracting the canopy temperature from the infrared canopy image.

16. The non-transient computer-readable storage medium according to claim 14 , wherein the determining the crop coefficient according to a canopy coverage ratio analyzed from the crop canopy image comprises:

inputting the crop canopy image into an analytical model, and obtaining the crop coefficient output from the analytical model, wherein

the analytical model is trained according to a sample crop canopy image and a canopy coverage ratio label and a crop coefficient label that are set in the sample crop canopy image; the analytical model comprises a ratio conversion layer and a ratio analysis layer;

the ratio conversion layer is configured to determine the canopy coverage ratio on the basis of the crop canopy image; and

the ratio analysis layer is configured to estimate the crop coefficient on the basis of the canopy coverage ratio.

17. The non-transient computer-readable storage medium according to claim 13 , wherein the determining a water stress state and a nutrient stress state on the basis of the crop coefficient and canopy temperature of the crop to be identified comprises:

setting a crop water stress index determined according to the canopy temperature as the water stress state; and

setting a normalized difference vegetation index determined according to the crop coefficient as the nutrient stress state.

18. The non-transient computer-readable storage medium according to claim 13 , wherein after the processing the water stress state and the nutrient stress state with a fuzzy control algorithm and determining a water and fertilizer stress decision result of the crop to be identified, the method further comprises:

controlling, in response to determining that the water and fertilizer stress decision result only comprises the irrigation amount, an irrigation apparatus to add water of the irrigation amount to the crop to be identified in the next cycle according to a first control instruction;

controlling, in response to determining that the water and fertilizer stress decision result only comprises the topdressing amount, a fertilization apparatus to add fertilizers of the topdressing amount to the crop to be identified in the next cycle according to a second control instruction; and

controlling, in response to determining that the water and fertilizer stress decision result only comprises the topdressing amount, an irrigation apparatus to add water of the irrigation amount to the crop to be identified in the next cycle according to a third control instruction, and a fertilization apparatus to add fertilizers of the topdressing amount to the crop to be identified in the next cycle.

19. A decision-making apparatus for water and fertilizer stress of crops, comprising:

an image processing module configured to determine a current crop coefficient and canopy temperature of a crop to be identified on the basis of a first crop image and a second crop image;

a water and fertilizer stress state identification module configured to determine a water stress state and a nutrient stress state on the basis of the current crop coefficient and canopy temperature of the crop to be identified; and

a water and fertilizer stress state decision-making module configured to process the water stress state and the nutrient stress state with a fuzzy control algorithm, and determine a water and fertilizer stress decision result of the crop to be identified, wherein

the decision result comprises an irrigation amount and/or a topdressing amount for the crop to be identified in a next cycle; the first crop image is a visible light image obtained by photographing the crop to be identified in full view with a terminal; the second crop image is a thermal infrared image obtained by photographing the crop to be identified in full view with an infrared module connected to the terminal; and

the water and fertilizer stress state decision-making module comprises a fuzzy control unit and a decision-making unit, wherein

the fuzzy control unit is configured to convert the water stress state and the nutrient stress state into linguistic variables, input the linguistic variables into a fuzzy controller, and obtain an irrigation instruction output from the fuzzy controller;

the decision-making unit is configured to set, in response to determining that first identifier information and second identifier information that are carried by the irrigation instruction are non-null and null respectively, an irrigation amount determined according to the crop coefficient and a reference crop evapotranspiration amount as the water and fertilizer stress decision result;

the decision-making unit is further configured to set, in response to determining that first identifier information and second identifier information that are carried by the irrigation instruction are null and non-null respectively, a topdressing amount determined according to the nutrient stress state and species information of the crop to be identified as the water and fertilizer stress decision result; and

the decision-making unit is further configured to set, in response to determining that neither of first identifier information and second identifier information that are carried by the irrigation instruction is null, an irrigation amount determined according to the crop coefficient and a reference crop evapotranspiration amount and a topdressing amount determined according to the nutrient stress state and species information of the crop to be identified as the water and fertilizer stress decision result, wherein

the fuzzy controller is created on the basis of a corresponding fuzzy rule between an input fuzzy set and an output fuzzy set; the input fuzzy set is determined according to a membership function set on the basis of fuzzy scales obtained by conducting fuzzy division on the water stress state and the nutrient stress state; the output fuzzy set is determined according to a membership function set on the basis of strategy scales of a plurality of water and fertilizer stress states obtained by clarifying a fuzzy language; the fuzzy rule is created by determining corresponding water and fertilizer stress states according to different water stress states and nutrient stress states; identifier information carried by the irrigation instruction is determined according to the water and fertilizer stress state; the first identifier information is configured to indicate whether to conduct irrigation; and the second identifier information is configured to indicate whether to conduct topdressing.

Priority Claims (1)

References Cited (12)