Urban indigenous plant leaf area index actual measurement system and method

Abstract

The invention relates to the technical field of data processing, and discloses an urban indigenous plant leaf area index actual measurement system and method, and the system comprises the steps: S1, a data collection module which is used for collecting leaf image data of urban indigenous plants; s2, the image processing module is used for preprocessing the collected leaf image data and extracting contour and area information of the plant leaves; s3, the leaf area index calculation module is used for calculating a leaf area index (LAI) according to the preprocessed leaf area data, and the leaf area index formula is # imgabs0 #; S4, the storage module is used for storing the collected image data, the processing result and the calculated leaf area index; and S5, a display module which is used for displaying the calculation result of the leaf area index and related image data in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a system and method for measuring leaf area index of urban native plants. Background Art

[0002] Leaf Area Index (LAI) is an important parameter in plant ecology. It is a key indicator for measuring plant canopy density and leaf area, and is widely used in ecosystem health monitoring, crop growth analysis, forest resource assessment and other fields. Traditional leaf area index measurement methods mainly rely on direct measurement of leaf area or indirect measurement using optical instruments. However, these methods have problems such as complex operation, long time consumption, high cost, and limited accuracy, especially in large-scale measurements and complex vegetation environments.

[0003] With the acceleration of urbanization, the importance of urban greening and vegetation monitoring has become increasingly prominent. As native plant groups that adapt to local climatic conditions, urban native plants are of great significance to the maintenance and sustainable development of the ecological environment. Therefore, how to accurately, efficiently and automatically measure and monitor the leaf area index of urban native plants has become a key technical problem in vegetation ecological monitoring and urban environmental management.

[0004] In the existing technology, the method of calculating leaf area index through image processing and remote sensing technology has gradually gained attention. However, most of the existing image processing methods rely on simple image segmentation algorithms, which are difficult to accurately segment the leaf area under complex backgrounds, resulting in errors in the measurement results. At the same time, the application of multi-temporal data processing and spatial distribution analysis of leaf area index in the existing technology is relatively limited and cannot meet the needs of accurate prediction and monitoring of large-scale areas. Therefore, there is an urgent need for a system that integrates modern image processing algorithms and spatial interpolation technology to automatically, efficiently and accurately complete the measurement and analysis of leaf area index of urban native plants. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and provide a system and method for measuring the leaf area index of urban native plants.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] The leaf area index measurement system of urban native plants comprises: S1, a data acquisition module, which is used to collect leaf image data of urban native plants;

[0008] S2, an image processing module, used to pre-process the collected leaf image data and extract the contour and area information of the plant leaves;

[0009] S3, leaf area index calculation module, is used to calculate the leaf area index (LAI) according to the preprocessed leaf area data. The leaf area index formula is:

[0010] S4, a storage module, used to store the collected image data, processing results and calculated leaf area index;

[0011] S5, a display module, used for displaying the calculation results of the leaf area index and related image data in real time.

[0012] Preferably, the data acquisition module S1 includes a combination of high-resolution camera equipment and multi-spectral sensors to obtain reflectance data of plants under different spectra, improve the clarity and resolution of leaf images, and ensure that the subsequent image processing module can accurately extract leaf information.

[0013] Preferably, the image processing module S2 includes:

[0014] S21, image enhancement unit, used to improve image quality, remove noise and interference, image enhancement uses Gaussian filtering to reduce noise;

[0015] S22, an edge detection unit, used to identify the edge contour of the plant leaf;

[0016] S23, an image segmentation unit, used for separating the leaves from the background.

[0017] Preferably, the image segmentation in the image processing module S2 realizes the segmentation of leaves and background based on a convolutional neural network (CNN);

[0018] The convolution layer in CNN uses convolution operation to extract the features of leaf images. The formula of convolution operation is:

[0019]

[0020] Among them, X is the input image tensor, W is the convolution kernel, Z is the output feature map, and b k is a bias term. The convolutional neural network (CNN) model uses a Softmax classifier to classify leaves;

[0021] The formula used by the Softmax classifier is:

[0022] Among them, z i is the score of category i, and P(y=i|X) is the probability that the leaf belongs to category i.

[0023] Preferably, the leaf area index calculation module S3 calculates the area of ​​each leaf based on the segmentation result of the leaf image, and accumulates these areas to calculate the total leaf area;

[0024] The area of ​​each leaf is calculated by multiplying the number of pixels in the segmented image by the actual area of ​​each pixel, using the formula:

[0025] Among them, pi is the number of pixels in the leaf area after segmentation, A pixcl is the actual area corresponding to each pixel.

[0026] Preferably, the system supports processing of multi-temporal images to monitor temporal changes in leaf area index;

[0027] The time series analysis model was used to analyze the temporal trend of leaf area index, and the autoregressive moving average model (ARMA) was adopted, and the formula was:

[0028] Among them, X t is the LAI value at a certain moment, εt is the error term, ai and β j are model parameters.

[0029] The system combines with a GPS module to collect the geographical coordinates of data points so as to generate a spatial distribution map of the leaf area index.

[0030] The urban native plant leaf area index measurement system according to claim 1 is characterized in that the display module of the system is used to display the spatial distribution diagram and time change curve of the leaf area index.

[0031] Preferably, the method for measuring the leaf area index of urban native plants comprises the following steps:

[0032] (1) collecting leaf image data through an image acquisition module;

[0033] (2) Use image processing modules to enhance, remove noise, detect edges, and segment images;

[0034] (3) Extract leaf edges and calculate area through convolutional neural network;

[0035] (4) Calculate the leaf area index according to the leaf area index formula;

[0036] (5) Analyze the temporal changes of LAI by combining multi-temporal data;

[0037] (6) Generate and display the spatial distribution map of LAI.

[0038] Compared with the prior art, the urban native plant leaf area index measurement system and method provided by the present invention has the following beneficial effects:

[0039] (1) Accurate leaf image segmentation and recognition: The present invention effectively solves the error problem of traditional image segmentation methods in complex backgrounds by introducing the convolutional neural network (CNN) algorithm. CNN can automatically learn and extract leaf features, thereby achieving high-precision leaf segmentation and improving the accuracy of leaf area index measurement.

[0040] (2) Vegetation recognition based on multispectral data: The present invention combines a multispectral sensor and calculates the normalized difference vegetation index (NDVI) to further enhance the recognition effect of leaf areas, avoid the uncertainty of a single optical image due to changes in illumination, and ensure the stability and accuracy of leaf area index calculation;

[0041] (3) Efficient spatiotemporal data processing and analysis: The system integrates time series analysis and spatial interpolation algorithms, such as Kriging interpolation and inverse distance weighted (IDW) interpolation, and can accurately predict the leaf area index and analyze its spatiotemporal distribution in large areas based on multi-temporal data. This function helps to monitor the growth status of urban native plants and their response to ecological and environmental changes over a long period of time.

[0042] (4) Automation and real-time display function: Through the highly automated design of data collection, processing and analysis, the system realizes the full process automation from image acquisition to leaf area index calculation. Combined with geographic information system (GIS) technology, the system can generate a spatial distribution map of the leaf area index and display the analysis results in real time, which is convenient for urban planning and environmental management departments to monitor the vegetation status in real time and provide decision support;

[0043] (5) Application in large-scale areas: The system and method of the present invention are particularly suitable for vegetation monitoring and analysis in large-scale urban areas. They can not only improve measurement efficiency, but also provide high-quality data support while reducing labor costs, thus meeting the needs of modern urban ecological environment management. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic diagram of the process structure of the urban native plant leaf area index measurement system and method of the present invention. DETAILED DESCRIPTION

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0046] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0047] like Figure 1 As shown, the urban native plant leaf area index measurement system provided by the embodiment of the present invention includes:

[0048] S1, data acquisition module, used to collect leaf image data of urban native plants;

[0049] S2, an image processing module, used to pre-process the collected leaf image data and extract the contour and area information of the plant leaves;

[0050] S3, leaf area index calculation module, is used to calculate the leaf area index (LAI) according to the preprocessed leaf area data. The leaf area index formula is:

[0051] S4, a storage module, used to store the collected image data, processing results and calculated leaf area index;

[0052] S5, a display module, used for displaying the calculation results of the leaf area index and related image data in real time.

[0053] The display module of the system is used to display the spatial distribution diagram and time variation curve of the leaf area index.

[0054] In one embodiment provided by the present invention, the data acquisition module S1 includes a combination of a high-resolution camera and a multi-spectral sensor to obtain reflection data of plants under different spectra, improve the clarity and resolution of leaf images, and ensure that the subsequent image processing module can accurately extract leaf information.

[0055] The image processing module S2 includes:

[0056] S21, image enhancement unit, used to improve image quality, remove noise and interference, image enhancement uses Gaussian filtering to reduce noise;

[0057] S22, an edge detection unit, used to identify the edge contour of the plant leaf;

[0058] S23, an image segmentation unit, used for separating the leaves from the background.

[0059] The image segmentation in the image processing module S2 realizes the segmentation of leaves and background based on convolutional neural network (CNN);

[0060] The convolution layer in CNN uses convolution operation to extract the features of leaf images. The formula of convolution operation is:

[0061]

[0062] Among them, X is the input image tensor, W is the convolution kernel, Z is the output feature map, and b k is a bias term. The convolutional neural network (CNN) model uses a Softmax classifier to classify leaves;

[0063] The formula used by the Softmax classifier is:

[0064] Among them, z i is the score of category i, and P(y=i|X) is the probability that the leaf belongs to category i.

[0065] The leaf area index calculation module S3 calculates the area of ​​each leaf based on the segmentation result of the leaf image, and accumulates these areas to calculate the total leaf area;

[0066] The area of ​​each leaf is calculated by multiplying the number of pixels in the segmented image by the actual area of ​​each pixel, using the formula:

[0067] Among them, pi is the number of pixels in the leaf area after segmentation, A pixcl is the actual area corresponding to each pixel.

[0068] The system supports processing of multi-temporal images to monitor the temporal changes of leaf area index;

[0069] The time series analysis model was used to analyze the temporal trend of leaf area index, and the autoregressive moving average model (ARMA) was adopted, and the formula was:

[0070] Among them, X t is the LAI value at a certain moment, εt is the error term, ai and β j are model parameters.

[0071] The system combines with a GPS module to collect the geographical coordinates of data points so as to generate a spatial distribution map of the leaf area index.

[0072] In another embodiment provided by the present invention, the method for measuring the leaf area index of urban native plants comprises the following steps:

[0073] (1) collecting leaf image data through an image acquisition module;

[0074] (2) Use image processing modules to enhance, remove noise, detect edges, and segment images;

[0075] (3) Extract leaf edges and calculate area through convolutional neural network;

[0076] (4) Calculate the leaf area index according to the leaf area index formula;

[0077] (5) Analyze the temporal changes of LAI by combining multi-temporal data;

[0078] (6) Generate and display the spatial distribution map of LAI.

[0079] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The urban native plant leaf area index measurement system is characterized by: The system comprises: S1, data acquisition module, used to collect leaf image data of urban native plants; S2, an image processing module, used to pre-process the collected leaf image data and extract the contour and area information of the plant leaves; S3, leaf area index calculation module, is used to calculate the leaf area index (LAI) according to the preprocessed leaf area data. The leaf area index formula is: S4, a storage module, used to store the collected image data, processing results and calculated leaf area index; S5, a display module, used for displaying the calculation results of the leaf area index and related image data in real time.

2. The urban native plant leaf area index measurement system according to claim 1, characterized in that: The data acquisition module S1 includes a combination of high-resolution camera equipment and multi-spectral sensors to obtain reflectance data of plants under different spectra, improve the clarity and resolution of leaf images, and ensure that the subsequent image processing module can accurately extract leaf information.

3. The urban native plant leaf area index measurement system and method according to claim 1, characterized in that: The image processing module S2 includes: S21, image enhancement unit, used to improve image quality, remove noise and interference, image enhancement uses Gaussian filtering to reduce noise; S22, an edge detection unit, used to identify the edge contour of the plant leaf; S23, an image segmentation unit, used for separating the leaves from the background.

4. The urban native plant leaf area index measurement system and method according to claim 1, characterized in that: The image segmentation in the image processing module S2 realizes the segmentation of leaves and background based on convolutional neural network (CNN); The convolution layer in CNN uses convolution operation to extract the features of leaf images. The formula of convolution operation is: Among them, X is the input image tensor, W is the convolution kernel, Z is the output feature map, and b k is the bias term.

5. The urban native plant leaf area index measurement system and method according to claim 1, characterized in that: The convolutional neural network (CNN) model uses a Softmax classifier to classify leaves; The formula used by the Softmax classifier is: Among them, z i is the score of category i, and P(y=i|X) is the probability that the leaf belongs to category i.

6. The urban native plant leaf area index measurement system and method according to claim 1, characterized in that: The leaf area index calculation module S3 calculates the area of ​​each leaf based on the segmentation result of the leaf image, and accumulates these areas to calculate the total leaf area; The area of ​​each leaf is calculated by multiplying the number of pixels in the segmented image by the actual area of ​​each pixel, using the formula: Among them, pi is the number of pixels in the leaf area after segmentation, A pixcl is the actual area corresponding to each pixel.

7. The urban native plant leaf area index measurement system and method according to claim 1, characterized in that: The system supports processing of multi-temporal images to monitor the temporal changes of leaf area index; The time series analysis model was used to analyze the temporal trend of leaf area index, and the autoregressive moving average model (ARMA) was adopted, and the formula was: Among them, X t is the LAI value at a certain moment, εt is the error term, ai and β j are model parameters.

8. The urban native plant leaf area index measurement system according to claim 1, characterized in that: The system combines with a GPS module to collect the geographical coordinates of data points so as to generate a spatial distribution map of the leaf area index.

9. The urban native plant leaf area index measurement system according to claim 1, characterized in that: The display module of the system is used to display the spatial distribution diagram and time variation curve of the leaf area index.

10. The method for measuring leaf area index of urban native plants according to any one of claims 1 to 9, characterized in that: The following steps are involved: (1) collecting leaf image data through an image acquisition module; (2) Use image processing modules to enhance, remove noise, detect edges, and segment images; (3) Extract leaf edges and calculate area through convolutional neural network; (4) Calculate the leaf area index according to the leaf area index formula; (5) Analyze the temporal changes of LAI by combining multi-temporal data; (6) Generate and display the spatial distribution map of LAI.

Images