LAI field measurement and analysis method and system for indigenous plants in urban residential area

Abstract

The invention relates to the field of ecological environment monitoring, and discloses an urban residential area indigenous plant LAI field measurement and analysis method and system, and the system comprises the steps: S1, an image collection module which is used for obtaining vertical and inclined plant canopy images of a target plant community; s2, a data processing module which is used for carrying out the preprocessing of the collected image, and the preprocessing comprises image denoising, gray level conversion and edge detection; s3, an LAI calculation module for calculating a leaf area index through a formula # imgabs0 #, and S4, an analysis and output module for displaying a calculation result in a chart form and providing a dynamic change trend; s5, a time sequence analysis module which predicts the change trend of the leaf area index by adopting an ARIMA model based on historical image data and time dimensions; and S6, a geographic information association module which combines a measurement result with a geographic information system (GIS) and is used for accurately evaluating and monitoring the urban ecological environment.

Description

Technical Field

[0001] The present invention relates to the field of ecological environment monitoring, and in particular to a method and system for field calculation and analysis of LAI of native plants in urban residential areas. Background Art

[0002] With the acceleration of urbanization, the urban ecological environment is facing increasingly severe challenges. Urban green spaces, especially native plants (local plants adapted to local climate and soil conditions), play an important role in air purification, temperature regulation, soil protection, and biodiversity maintenance in the urban environment. Therefore, the health status assessment and long-term monitoring of urban ecosystems have become increasingly important.

[0003] Leaf area index (LAI) is a key indicator for characterizing plant canopy structure and density, usually defined as the total area of ​​plant leaves per unit surface area. LAI can reflect plant growth status, photosynthesis efficiency, transpiration intensity, etc., and is one of the important parameters for assessing ecosystem health and productivity.

[0004] However, traditional LAI calculation methods mainly rely on manual measurement or indirect estimation based on theoretical models, which has the following disadvantages: limited accuracy: manual measurement has large errors and cannot fully reflect the actual growth of plants in large-scale green spaces in cities; low efficiency: manual operation is time-consuming and labor-intensive, especially in large green spaces or complex plant communities, where measurement work is difficult to complete efficiently; insufficient spatiotemporal coverage: traditional measurement methods cannot dynamically track plant growth changes and lack detailed monitoring of different spatial and temporal scales.

[0005] In recent years, with the rapid development of image processing technology and remote sensing technology, image-based LAI calculation methods have gradually received attention. Through multispectral cameras, drones or satellite remote sensing technologies, a large amount of plant canopy image data can be collected. Using these data combined with advanced image processing algorithms and mathematical models, LAI can be calculated more accurately and efficiently. However, existing image processing methods and LAI calculation models still face some problems when dealing with complex urban green spaces, such as:

[0006] The image processing algorithm is not perfect: when dealing with complex plant communities and ambient light conditions, the existing image algorithm cannot accurately extract effective plant canopy information, resulting in large deviations in the measurement results.

[0007] Insufficient use of multispectral data: Spectral data in different bands can provide rich plant information, especially when assessing plant health, but existing systems fail to fully utilize these data.

[0008] Therefore, there is an urgent need for a system that integrates image processing, multispectral data analysis, and machine learning technologies to efficiently and accurately measure and analyze the leaf area index (LAI) of native plants in urban residential areas, providing a scientific basis for the management and maintenance of urban ecosystems. Summary of the invention

[0009] The purpose of the present invention is to solve the shortcomings of the prior art and provide a method and system for field measurement and analysis of LAI of native plants in urban residential areas.

[0010] In order to achieve the above object, the present invention provides the following technical solutions:

[0011] A field measurement and analysis system for LAI of native plants in urban residential areas, the system comprising: S1, an image acquisition module, for acquiring plant canopy images at vertical and oblique angles of a target plant community;

[0012] S2, data processing module, used for preprocessing the collected images, including image denoising, grayscale conversion and edge detection;

[0013] S3, LAI calculation module, through the formula Calculate the leaf area index, where: I0 is the incident light intensity, I is the penetrating light intensity, and K is an optical constant, which is determined by plant species and environmental factors;

[0014] S4, analysis and output module, displays the calculation results in the form of charts and provides dynamic change trends;

[0015] S5, time series analysis module, based on historical image data and time dimension, uses ARIMA model to predict the change trend of leaf area index;

[0016] S6. Geographic Information Association Module: Combines the measurement results with the Geographic Information System (GIS) for accurate assessment and monitoring of the urban ecological environment.

[0017] Preferably, the image acquisition module S1 includes:

[0018] S11, multispectral camera, used to obtain plant images in different bands to provide additional analysis of plant health;

[0019] S12, GPS module, is used to mark the geographic location data of image acquisition for spatial analysis and long-term monitoring.

[0020] Preferably, the data processing module S2 includes:

[0021] S21, a binarization processing unit, used for binarizing the grayscale image and extracting the plant canopy contour;

[0022] S22, image segmentation unit, uses an image segmentation method based on k-means clustering to distinguish between background and plant areas and improve the accuracy of LAI calculation.

[0023] Preferably, the LAI calculation module S3 adopts an improved algorithm:

[0024] Based on the improved Belfast light scattering model, taking into account the multiple scattering of light between leaves, the calculation formula is: Where θ is the incident light angle and f(θ) is the scattering correction factor.

[0025] The data processing module S2 adopts the following algorithm:

[0026] An image processing method based on morphological operations, combining dilation, erosion, and opening and closing operations, is used to separate leaves from the background. An adaptive threshold method is used to perform binarization processing on images under different lighting conditions.

[0027] Preferably, the image feature extraction module uses a convolutional neural network (CNN) to extract features from plant canopy images;

[0028] The leaf area prediction module estimates LAI based on the extracted features using the trained regression model. The estimation formula is: Where wi is the weight of the model, f i (x) is the feature of the input image.

[0029] The neural network training module includes the following steps:

[0030] Collect multiple sets of annotated plant canopy image data;

[0031] The neural network was trained using a cross-validation method to ensure the accuracy and robustness of the LAI prediction results.

[0032] It also includes a geographic information association module: combining measurement results with the Geographic Information System (GIS) for accurate assessment and monitoring of the urban ecological environment

[0033] Preferably, the method for field measurement and analysis of LAI of native plants in urban residential areas comprises the following steps:

[0034] (1) Image acquisition step: acquiring images of plant canopies through a multi-angle image acquisition device;

[0035] (2) Data processing steps: preprocessing the collected images, including denoising, grayscale conversion, and edge detection;

[0036] (3) LAI calculation steps: Calculate leaf area index;

[0037] (4) Data output step: output the LAI calculation results in graphical and numerical form;

[0038] (5) Time series analysis step: Based on historical image data and time dimension, the ARIMA model is used to predict the change trend of leaf area index;

[0039] (6) Geographic information association step: Combine the measurement results with the geographic information system (GIS) for accurate assessment and monitoring of the urban ecological environment.

[0040] Compared with the prior art, the method and system for field measurement and analysis of LAI of native plants in urban residential areas provided by the present invention have the following beneficial effects:

[0041] (1) High measurement accuracy:

[0042] The present invention uses a multispectral camera to collect plant canopy images from multiple angles. Combined with multispectral data, it can accurately distinguish different plant species and health conditions, ensuring the accuracy of LAI calculation. Based on the improved optical scattering model and convolutional neural network (CNN) algorithm, it can correct the illumination changes in complex environments, making the measurement results more reliable. The system uses advanced image processing algorithms, such as Gaussian filtering and Canny edge detection, which can effectively extract the contour information of the plant canopy and improve the accuracy of LAI calculation.

[0043] (2) High calculation efficiency:

[0044] Compared with traditional manual measurement or theoretical models, the present invention greatly improves the measurement efficiency through image acquisition and automatic calculation. The collected image data can be quickly processed by the system to obtain the LAI value in real time, which greatly reduces the need for manual intervention. The system can quickly measure a large range of urban green spaces and is suitable for complex and diverse urban plant communities, especially in large urban parks or green belts, which can significantly improve the efficiency of data acquisition;

[0045] (3) Dynamic monitoring and prediction function:

[0046] The present invention can conduct long-term dynamic monitoring. Through the time series analysis function, the system can record and analyze the LAI change trend of plant communities, providing predictive data support for the maintenance and management of urban ecosystems. Combined with the geographic information system (GIS), it can generate a spatial distribution map of LAI and conduct comparative analysis at different time nodes to help managers understand the health changes of urban green spaces.

[0047] (4) Wide range of applications:

[0048] The system is suitable for various urban ecological environments, including green areas such as parks, residential areas, and commercial areas. Through flexible collection equipment (such as handheld devices, drones, etc.), the system can adapt to the needs of different scenarios. The system can be applied to a variety of plant types, especially the ecological monitoring of native plants, to help urban managers better protect and develop local species;

[0049] (5) Automation and intelligence:

[0050] The system integrates machine learning technologies such as convolutional neural networks. By training on large amounts of image data, it can continuously optimize the calculation results of LAI. It has self-learning and self-optimization functions to reduce human errors. The system can automatically handle the entire process from data collection to output, greatly simplifying the complexity of the measurement work and improving the convenience of operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the process structure of the on-site measurement and analysis method and system of LAI of native plants in urban residential areas of the present invention. DETAILED DESCRIPTION

[0052] 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.

[0053] 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.

[0054] like Figure 1 As shown, the embodiment of the present invention provides a system for field measurement and analysis of LAI of native plants in urban residential areas, the system comprising: S1, an image acquisition module, for acquiring plant canopy images at vertical and oblique angles of a target plant community;

[0055] S2, data processing module, used for preprocessing the collected images, including image denoising, grayscale conversion and edge detection;

[0056] S3, LAI calculation module, through the formula Calculate the leaf area index, where: I0 is the incident light intensity, I is the penetrating light intensity, and K is an optical constant, which is determined by plant species and environmental factors;

[0057] S4, analysis and output module, displays the calculation results in the form of charts and provides dynamic change trends;

[0058] S5, time series analysis module, based on historical image data and time dimension, uses ARIMA model to predict the change trend of leaf area index;

[0059] S6. Geographic Information Association Module: Combines the measurement results with the Geographic Information System (GIS) for accurate assessment and monitoring of the urban ecological environment.

[0060] In one embodiment provided by the present invention, the image acquisition module S1 includes:

[0061] S11, multispectral camera, used to obtain plant images in different bands to provide additional analysis of plant health;

[0062] S12, GPS module, is used to mark the geographic location data of image acquisition for spatial analysis and long-term monitoring.

[0063] The data processing module S2 includes:

[0064] S21, a binarization processing unit, used for binarizing the grayscale image and extracting the plant canopy contour;

[0065] S22, image segmentation unit, uses an image segmentation method based on k-means clustering to distinguish between background and plant areas and improve the accuracy of LAI calculation.

[0066] The LAI calculation module S3 adopts an improved algorithm:

[0067] Based on the improved Belfast light scattering model, taking into account the multiple scattering of light between leaves, the calculation formula is: Where θ is the incident light angle and f(θ) is the scattering correction factor.

[0068] The data processing module S2 adopts the following algorithm:

[0069] An image processing method based on morphological operations, combining dilation, erosion, and opening and closing operations, is used to separate leaves from the background. An adaptive threshold method is used to perform binarization processing on images under different lighting conditions.

[0070] The image feature extraction module uses a convolutional neural network (CNN) to extract features from plant canopy images;

[0071] The leaf area prediction module estimates LAI based on the extracted features using the trained regression model. The estimation formula is: Where wi is the weight of the model, f i (x) is the feature of the input image.

[0072] The neural network training module includes the following steps:

[0073] Collect multiple sets of annotated plant canopy image data;

[0074] The neural network was trained using a cross-validation method to ensure the accuracy and robustness of the LAI prediction results.

[0075] It also includes a geographic information association module: combining measurement results with the Geographic Information System (GIS) for accurate assessment and monitoring of the urban ecological environment

[0076] Preferably, the method for field measurement and analysis of LAI of native plants in urban residential areas comprises the following steps:

[0077] (1) Image acquisition step: acquiring images of plant canopies through a multi-angle image acquisition device;

[0078] (2) Data processing steps: preprocessing the collected images, including denoising, grayscale conversion, and edge detection;

[0079] (3) LAI calculation steps: Calculate leaf area index;

[0080] (4) Data output step: output the LAI calculation results in graphical and numerical form;

[0081] (5) Time series analysis step: Based on historical image data and time dimension, the ARIMA model is used to predict the change trend of leaf area index;

[0082] (6) Geographic information association step: Combine the measurement results with the geographic information system (GIS) for accurate assessment and monitoring of the urban ecological environment.

[0083] 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 field measurement and analysis system of LAI of native plants in urban residential areas is characterized by: The system comprises: S1, an image acquisition module, for acquiring plant canopy images at vertical and oblique angles of a target plant community; S2, data processing module, used for preprocessing the collected images, including image denoising, grayscale conversion and edge detection; S3, LAI calculation module, through the formula Calculate the leaf area index, where: I0 is the incident light intensity, I is the penetrating light intensity, and K is an optical constant, which is determined by plant species and environmental factors; S4, analysis and output module, displays the calculation results in the form of charts and provides dynamic change trends; S5, time series analysis module, based on historical image data and time dimension, uses ARIMA model to predict the change trend of leaf area index; S6. Geographic Information Association Module: Combines the measurement results with the Geographic Information System (GIS) for accurate assessment and monitoring of the urban ecological environment.

2. The system for measuring and analyzing LAI of native plants in urban residential areas according to claim 1 is characterized in that: The image acquisition module S1 includes: S11, multispectral camera, used to obtain plant images in different bands to provide additional analysis of plant health; S12, GPS module, is used to mark the geographic location data of image acquisition for spatial analysis and long-term monitoring.

3. The method and system for field calculation and analysis of LAI of native plants in urban residential areas according to claim 1 is characterized in that: The data processing module S2 includes: S21, a binarization processing unit, used for binarizing the grayscale image and extracting the plant canopy contour; S22, image segmentation unit, uses an image segmentation method based on k-means clustering to distinguish between background and plant areas and improve the accuracy of LAI calculation.

4. The method and system for field calculation and analysis of LAI of native plants in urban residential areas according to claim 1 is characterized in that: The LAI calculation module S3 adopts an improved algorithm: Based on the improved Belfast light scattering model, taking into account the multiple scattering of light between leaves, the calculation formula is: Where θ is the incident light angle and f(θ) is the scattering correction factor.

5. The method and system for field calculation and analysis of LAI of native plants in urban residential areas according to claim 1 is characterized in that: The data processing module S2 adopts the following algorithm: An image processing method based on morphological operations, combining dilation, erosion, and opening and closing operations, is used to separate leaves from the background. An adaptive threshold method is used to perform binarization processing on images under different lighting conditions.

6. The method and system for field calculation and analysis of LAI of native plants in urban residential areas according to claim 1 is characterized in that: The image feature extraction module uses a convolutional neural network (CNN) to extract features from plant canopy images; The leaf area prediction module estimates LAI based on the extracted features using the trained regression model. The estimation formula is: Where wi is the weight of the model, f i (x) is the feature of the input image.

7. The system for field calculation and analysis of LAI of native plants in urban residential areas according to claim 1 is characterized in that: The neural network training module includes the following steps: Collect multiple sets of annotated plant canopy image data; The neural network was trained using a cross-validation method to ensure the accuracy and robustness of the LAI prediction results.

8. The method and system for field calculation and analysis of LAI of native plants in urban residential areas according to claim 1 is characterized in that: It also includes a geographic information association module: combining measurement results with the Geographic Information System (GIS) for accurate assessment and monitoring of the urban ecological environment.

9. The method for measuring leaf area index of urban native plants according to any one of claims 1 to 8, characterized in that: The following steps are involved: (1) Image acquisition step: acquiring images of plant canopies through a multi-angle image acquisition device; (2) Data processing steps: preprocessing the collected images, including denoising, grayscale conversion, and edge detection; (3) LAI calculation steps: Calculate leaf area index; (4) Data output step: output the LAI calculation results in graphical and numerical form; (5) Time series analysis step: Based on historical image data and time dimension, the ARIMA model is used to predict the change trend of leaf area index; (6) Geographic information association step: Combine the measurement results with the geographic information system (GIS) for accurate assessment and monitoring of the urban ecological environment.

Images