Gramineae crop coverage rate measuring method

Abstract

The invention discloses a gramineous crop coverage rate measurement method. The method comprises the following steps: step 1, acquiring farmland plot image data; 2, preprocessing the farmland plot image data; 3, calculating a green phase index of the preprocessed farmland plot image data; 4, determining an optimal green phase index threshold value; 5, calculating the coverage rate of gramineous crops in the farmland plot according to the green phase index and a green phase index threshold value; compared with a traditional manual visual inspection method, automatic and high-precision measurement of the crop coverage rate is achieved, the measurement efficiency and accuracy are greatly improved, and the gramineous crop coverage rate measurement method reduces the human input of manual measurement and reduces the agricultural production cost; the crop coverage rate change is monitored in real time, timely and accurate growth information is provided for agricultural production, a scientific basis is provided for agricultural management and decision making, and management measures such as precise irrigation and fertilization can be achieved.

Description

Technical Field

[0001] The invention relates to the field of agricultural information technology, and in particular to a method for measuring the coverage rate of gramineous crops. Background Art

[0002] Crop coverage rate is a key indicator for evaluating crop growth status, and is of great significance for precision agricultural management, crop yield prediction and variety improvement. Traditional manual visual inspection methods are not only time-consuming and laborious, but also easily affected by subjective factors, resulting in poor accuracy and consistency of measurement results. With the rapid development of remote sensing technology and computer vision technology, image-based crop coverage rate measurement methods have gradually become a research hotspot. However, due to the unique leaf morphology and dense growth characteristics of grass crops, existing image recognition algorithms still face challenges in accurately distinguishing crops from soil, weeds, etc. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention provides a method for measuring the coverage rate of gramineous crops, which solves the problem that the traditional manual visual inspection method is time-consuming and labor-intensive, easily affected by subjective factors, and leads to poor accuracy and consistency of measurement results.

[0004] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0005] A method for measuring grass crop coverage is provided, comprising:

[0006] Step 1: Obtain farmland plot image data;

[0007] Step 2: preprocessing the farmland plot image data;

[0008] Step 3, calculating the green phase index of the preprocessed farmland plot image data;

[0009] Step 4: Determine the optimal green phase index threshold;

[0010] Step 5: Calculate the coverage rate of gramineous crops in the farmland plot according to the green phase index and the green phase index threshold.

[0011] Further, in step 1, a drone equipped with a standard RGB camera is operated to cover a farmland plot according to a predetermined flight path and height, and the drone captures images of the farmland plot multiple times during the flight, and obtains image data of the farmland plot under different lighting conditions, which helps to more comprehensively evaluate the coverage of crops in the farmland plot.

[0012] Furthermore, in step 2, the preprocessing method of the farmland plot image data is as follows: preliminarily screening the farmland plot image data to eliminate images of poor quality due to weather or equipment problems; digitally processing the remaining valid image data, using Gaussian filtering to reduce the noise of the valid image data, smoothing the differences between pixels, and then filtering the pixel points of the valid image data according to the pixel brightness, selecting the pixel intensity area from the 2.5th percentile to the 97.5th percentile in the valid image, excluding the overexposed and underexposed parts in the valid image, reducing the impact on vegetation cover judgment, and improving the accuracy of the farmland plot image data.

[0013] Further, in step 3, the green phase index is used to evaluate the greenness of vegetation. The green phase index is designed based on the principle of photosynthesis by calculating the light intensity ratio of the green and red channels in the image. Because green plants absorb red light in visible light and reflect green light during photosynthesis, the intensity ratio of these two colors is measured to provide information about vegetation coverage and health status. The green phase index is obtained by converting the inverse cosine formula and is mainly used to evaluate the greenness of vegetation pixels. This index is particularly suitable for analyzing crop coverage. Specifically, the calculation formula of the green phase index is:

[0014]

[0015] Among them, GPVI is the green phase index, g is the intensity of the green channel of each pixel in the farmland plot image; r is the intensity of the red channel of each pixel in the farmland plot image; by performing trigonometric function processing on the ratio of green and red light intensity, a color relative angle value is calculated, which can reflect the green degree of vegetation. When the vegetation pixel is greener, the value of g is higher relative to r, and a lower GPVI angle value is obtained. Compared with the linear method, the inverse cosine function can amplify the difference in pixel color values ​​between green and red. Compared with linear smooth mapping, this method can improve the stability of subsequent threshold methods.

[0016] Further, in step 4, the green phase index threshold is used to distinguish between plant areas and non-plant areas in the farmland plot image, and the method for determining the green phase index threshold includes:

[0017] Step 4.1, calculating GPVI values ​​for all pixels of the entire farmland plot image data to form a distribution of GPVI values;

[0018] Step 4.2, traverse all possible green phase index thresholds, and for each green phase index threshold, divide the GPVI value into two groups of data, which are greater than the green phase index threshold and which are less than the green phase index threshold;

[0019] Step 4.3, calculate the internal variance and inter-group variance of the two groups of GPVI value data;

[0020] Step 4.4, select the green phase index threshold that maximizes the inter-group variance of the two groups of GPVI value data as the optimal green phase index threshold. The optimal green phase index threshold can maximize the distinction between vegetation and non-vegetation areas, because the maximized inter-group variance means that the distribution difference between the two groups is the largest.

[0021] Furthermore, in step 5, the method for calculating the coverage rate of gramineous crops in the farmland plot includes:

[0022] Step 5.1, calculate the plant coverage rate of each image in the preprocessed farmland plot image data: apply the determined threshold value to classify each pixel in the image, determine whether these pixels belong to plants, and compare the number of pixels determined to be plants with the total number of pixels in the image after removing the over- and under-exposed parts, so as to calculate the plant coverage rate of each image;

[0023] Step 5.2: Summarize the plant coverage of each image and calculate the average value to obtain the coverage of gramineous crops in the farmland. By comparing the coverage data at different time points, the growth dynamics and trends of crops can be analyzed, providing a scientific basis for agricultural management and decision-making.

[0024] The beneficial effects of the present invention are as follows: compared with the traditional manual visual method, the method for measuring the coverage rate of grass crops in this scheme realizes the automated and high-precision measurement of crop coverage rate, greatly improves the measurement efficiency and accuracy, reduces the manpower input of manual measurement, and reduces the cost of agricultural production; by real-time monitoring of changes in crop coverage rate, timely and accurate growth information is provided for agricultural production, a scientific basis is provided for agricultural management and decision-making, and it is helpful to realize management measures such as precise irrigation and fertilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The figure is a principle block diagram of a method for measuring grass crop coverage. DETAILED DESCRIPTION

[0026] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0027] like Figure 1As shown, a method for measuring the coverage rate of gramineous crops includes: step 1, obtaining image data of farmland plots; specifically, in step 1, operating a drone equipped with a standard RGB camera to cover the farmland plot according to a predetermined flight path and height, the drone captures images of the farmland plot multiple times during the flight, and obtains image data of the farmland plot under different lighting conditions, which helps to more comprehensively evaluate the coverage of crops in the farmland plots.

[0028] Step 2, preprocess the farmland plot image data; the farmland plot image data preprocessing method is: preliminarily screen the farmland plot image data to eliminate images of poor quality due to weather or equipment problems; digitally process the remaining valid image data, use Gaussian filtering to reduce the noise of the valid image data, smooth the differences between pixels, and then filter the pixel points of the valid image data according to the pixel brightness, select the pixel intensity area from the 2.5th percentile to the 97.5th percentile in the valid image, exclude the over-exposed and under-exposed parts in the valid image, reduce the impact on vegetation cover judgment, and improve the accuracy of the farmland plot image data.

[0029] Step 3: Calculate the green phase index of the preprocessed farmland image data; the green phase index is used to evaluate the greenness of vegetation. The green phase index is designed based on the principle of photosynthesis by calculating the light intensity ratio of the green and red channels in the image. Because green plants absorb red light in visible light and reflect green light during photosynthesis, it provides information about vegetation coverage and health by measuring the intensity ratio of these two colors. The green phase index is obtained by converting the inverse cosine formula and is mainly used to evaluate the greenness of vegetation pixels. This index is particularly suitable for analyzing crop coverage. Specifically, the calculation formula of the green phase index is:

[0030]

[0031] Among them, GPVI is the green phase index, g is the intensity of the green channel of each pixel in the farmland plot image; r is the intensity of the red channel of each pixel in the farmland plot image; by performing trigonometric function processing on the ratio of green and red light intensity, a color relative angle value is calculated, which can reflect the green degree of vegetation. When the vegetation pixel is greener, the value of g is higher relative to r, and a lower GPVI angle value is obtained. Compared with the linear method, the inverse cosine function can amplify the difference in pixel color values ​​between green and red. Compared with linear smooth mapping, this method can improve the stability of subsequent threshold methods.

[0032] Step 4: determine the optimal green phase index threshold; the green phase index threshold is used to distinguish between plant areas and non-plant areas in the farmland plot image. The method for determining the green phase index threshold includes:

[0033] Step 4.1, calculating GPVI values ​​for all pixels of the entire farmland plot image data to form a distribution of GPVI values;

[0034] Step 4.2, traverse all possible green phase index thresholds, and for each green phase index threshold, divide the GPVI value into two groups of data, which are greater than the green phase index threshold and which are less than the green phase index threshold;

[0035] Step 4.3, calculate the internal variance and inter-group variance of the two groups of GPVI value data;

[0036] Step 4.4, select the green phase index threshold that maximizes the inter-group variance of the two groups of GPVI value data as the optimal green phase index threshold. The optimal green phase index threshold can maximize the distinction between vegetation and non-vegetation areas, because the maximized inter-group variance means that the distribution difference between the two groups is the largest.

[0037] Step 5: Calculate the coverage rate of gramineous crops in the farmland plot according to the green phase index and the green phase index threshold. The method for calculating the coverage rate of gramineous crops in the farmland plot includes:

[0038] Step 5.1, calculate the plant coverage rate of each image in the preprocessed farmland plot image data: apply the determined threshold value to classify each pixel in the image, determine whether these pixels belong to plants, and compare the number of pixels determined to be plants with the total number of pixels in the image after removing the over- and under-exposed parts, so as to calculate the plant coverage rate of each image;

[0039] Step 5.2: Summarize the plant coverage of each image and calculate the average value to obtain the coverage of gramineous crops in the farmland. By comparing the coverage data at different time points, the growth dynamics and trends of crops can be analyzed, providing a scientific basis for agricultural management and decision-making.

[0040] In summary, compared with the traditional manual visual method, the method for measuring the coverage rate of grass crops in this scheme realizes the automated and high-precision measurement of crop coverage rate, greatly improves the measurement efficiency and accuracy, reduces the manpower input of manual measurement, and reduces the cost of agricultural production; by real-time monitoring of crop coverage changes, it provides timely and accurate growth information for agricultural production, provides a scientific basis for agricultural management and decision-making, and helps to implement precise irrigation, fertilization and other management measures.

Claims

1. A method for measuring the coverage rate of gramineous crops, characterized in that: include Step 1: Obtain farmland plot image data; Step 2: preprocessing the farmland plot image data; Step 3, calculating the green phase index of the preprocessed farmland plot image data; Step 4: Determine the optimal green phase index threshold; Step 5: Calculate the coverage rate of gramineous crops in the farmland plot according to the green phase index and the green phase index threshold.

2. The method for measuring the coverage rate of gramineous crops according to claim 1, characterized in that: In step 1, a drone equipped with a standard RGB camera is operated to cover a farmland plot according to a predetermined flight path and height. The drone captures images of the farmland plot multiple times during the flight and obtains image data of the farmland plot under different lighting conditions.

3. The method for measuring the coverage rate of gramineous crops according to claim 1, characterized in that: In step 2, the preprocessing method of the farmland plot image data is as follows: preliminarily screen the farmland plot image data to eliminate images of poor quality due to weather or equipment problems; digitally process the remaining valid image data, use Gaussian filtering to reduce the noise of the valid image data, smooth the differences between pixels, and then filter the pixels of the valid image data according to pixel brightness, select the pixel intensity area from the 2.5th percentile to the 97.5th percentile in the valid image, and exclude the overexposed and underexposed parts in the valid image.

4. The method for measuring the coverage rate of gramineous crops according to claim 3, characterized in that: In step 3, the green phase index is used to evaluate the greenness of vegetation. The calculation formula of the green phase index is: Among them, GPVI is the green phase index, g is the intensity of the green channel of each pixel in the farmland plot image; r is the intensity of the red channel of each pixel in the farmland plot image.

5. The method for measuring the coverage rate of gramineous crops according to claim 4, characterized in that: In step 4, the green phase index threshold is used to distinguish between plant areas and non-plant areas in the farmland plot image. The method for determining the green phase index threshold includes: Step 4.1, calculating GPVI values ​​for all pixels of the entire farmland plot image data to form a distribution of GPVI values; Step 4.2, traverse all possible green phase index thresholds, and for each green phase index threshold, divide the GPVI value into two groups of data, which are greater than the green phase index threshold and which are less than the green phase index threshold; Step 4.3, calculate the internal variance and inter-group variance of the two groups of GPVI value data; Step 4.4: Select the green phase index threshold that maximizes the inter-group variance of the two groups of GPVI value data as the optimal green phase index threshold.

6. The method for measuring the coverage rate of gramineous crops according to claim 5, characterized in that: In step 5, the method for calculating the coverage rate of gramineous crops in the farmland plot includes: Step 5.1, calculate the plant coverage rate of each image in the preprocessed farmland plot image data: apply the determined threshold value to classify each pixel in the image, determine whether these pixels belong to plants, and compare the number of pixels determined to be plants with the total number of pixels in the image after removing the over- and under-exposed parts, so as to calculate the plant coverage rate of each image; Step 5.2: Summarize the plant coverage calculated for each image and calculate the average value to obtain the grass crop coverage in the farmland plot.

Images