Garden landscape design system and method based on computer vision

Abstract

The application discloses a computer vision-based garden landscape design system and method, and relates to the technical field of computer vision, which comprises the following steps: receiving plant variety data, environmental parameter data and night spectrum distribution data of a night lighting area, extracting corresponding spectral sensitivity functions according to the plant variety data, combining the environmental parameter data and the night spectrum distribution data to calculate a light influence factor, identifying the current growth stage of the plants in the night lighting area if the light influence factor exceeds a preset threshold, extracting corresponding spectral target curves, combining the night spectrum distribution data and the spectral sensitivity functions to calculate a comprehensive spectral mismatch index, and finally obtaining a target output spectrum for guiding light adjustment in the area by optimization solution under the condition of meeting the preset visual brightness and change rate constraints; the beneficial effects are that the influence of night light on the photoperiod of plants can be quantitatively evaluated, the risk of ecological disturbance is reduced, and the landscape effect is dynamically optimized.

Description

Technical Field

[0001] This invention relates to the field of computer vision technology, and in particular to a computer vision-based landscape design system and method. Background Technology

[0002] As an important part of environmental art, landscape design aims to create spaces that coexist harmoniously with nature and enhance the aesthetic value and ecological function of cities. With the acceleration of urbanization, modern landscape design not only focuses on shaping daytime landscapes but also attaches great importance to the expressiveness and artistic effect of nighttime landscapes. Night lighting, as a key means to enrich the visual layers of gardens and enhance the viewing experience, has become an important part of landscape design.

[0003] However, existing garden nightscape designs often rely on the designer's experience to arrange and set the intensity of lighting. While this may enhance the artistic effect of the landscape to some extent, it often overlooks the differences in light requirements of plants at different growth stages. Continuous or unreasonable nighttime lighting can easily interfere with the natural photocycle of plants, affecting their normal growth and flowering, thereby weakening the ecological health and overall aesthetics of the garden.

[0004] Therefore, a computer vision-based landscape design system and method are proposed. Summary of the Invention

[0005] In view of the above-mentioned prior art, this application is hereby filed. Embodiments of this application provide a computer vision-based landscape design system and method, which can quantitatively assess the impact of nighttime light on plant photoperiod, reduce ecological disturbance risks, and dynamically optimize landscape effects.

[0006] According to one aspect of this application, a computer vision-based landscape design method is provided, comprising: receiving plant variety data, environmental parameter data, and nighttime spectral distribution data for each nighttime lighting area within a target garden; extracting corresponding spectral sensitivity functions based on the plant variety data and a preset plant spectral sensitivity function library; calculating a light influence factor for the degree of light interference of each nighttime lighting area on the photoperiod of plants within the area based on the environmental parameter data and the nighttime spectral distribution data; determining whether the light influence factor exceeds a preset influence threshold, and if so: acquiring image data of the nighttime lighting area and identifying the current growth stage of each plant variety within the area; extracting the target spectral curve of the corresponding plant variety from a preset spectral curve database based on the current growth stage; calculating a comprehensive spectral mismatch index within the nighttime lighting area based on the target spectral curve, the nighttime spectral distribution data, and the spectral sensitivity function; and, based on the comprehensive spectral mismatch index, solving and optimizing a target output spectrum for guiding the lighting adjustment of the nighttime lighting area under the condition of satisfying preset visual brightness and rate of change constraints.

[0007] According to another aspect of this application, a computer vision-based landscape design system is provided, comprising: a data acquisition module for receiving plant variety data, environmental parameter data, and nighttime spectral distribution data for each nighttime lighting area within a target garden; a function extraction module for extracting corresponding spectral sensitivity functions based on the plant variety data and a preset plant spectral sensitivity function library; a first calculation module for calculating a light impact factor on the degree of light interference of each nighttime lighting area on the photoperiod of plants within the area, based on the environmental parameter data and the nighttime spectral distribution data; a judgment module for judging whether the light impact factor exceeds a preset impact threshold; and an image recognition module. The system comprises three modules: a first module for acquiring image data of the nighttime illumination area and identifying the current growth stage of each plant species within the area; a second module for extracting the target spectral curve of the corresponding plant species from a preset spectral curve database based on the current growth stage; a third module for calculating the comprehensive spectral mismatch index within the nighttime illumination area based on the target spectral curve, the nighttime spectral distribution data, and the spectral sensitivity function; and a fourth module for optimizing the target output spectrum to guide the adjustment of the nighttime illumination area based on the comprehensive spectral mismatch index, under the condition of satisfying preset visual brightness and rate of change constraints.

[0008] According to another aspect of this application, an electronic device is provided, including a memory and a processor, the memory being used to store computer-executable instructions, and the processor being used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the method described above.

[0009] According to another aspect of this application, a computer storage medium is provided that stores computer-executable instructions thereon, which, when executed by a processor, implement the steps of the method described above.

[0010] Compared with the prior art, the computer vision-based landscape design system and method according to the embodiments of this application can dynamically optimize the nighttime lighting spectrum distribution by quantitatively analyzing the influence of plant spectral sensitivity and environmental parameters on photoperiod. This reduces interference with plant growth while ensuring the landscape effect, and has the advantages of quantitatively assessing the impact of nighttime light on plant photoperiod, reducing the risk of ecological disturbance, and dynamically optimizing the landscape effect. Attached Figure Description

[0011] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0012] Figure 1 This is a flowchart of the computer vision-based landscape design method of the present invention.

[0013] Figure 2 This is a block diagram of the computer vision-based landscape design system of the present invention.

[0014] Figure 3 This is a block diagram of an electronic device according to the present invention. Detailed Implementation

[0015] Hereinafter, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0016] Application Overview

[0017] In current garden night lighting design, the setting of lighting parameters relies on human experience and lacks quantitative analysis of the sensitivity of plant photoperiod. This can easily lead to light interference exceeding the plant's tolerance threshold, resulting in a mismatch between spectral energy distribution and the plant's photomorphogenesis requirements.

[0018] If the above problems are not addressed, the persistent spectral energy mismatch will lead to disordered plant photomorphogenesis, manifested as delayed flowering, inhibition of chlorophyll synthesis, and exacerbation of photoinhibition effects. In mixed planting scenarios with multiple species, the cumulative effect of spectral interference may cause an imbalance in the ecological niche competition of plant communities, accelerating the degradation of vegetation in local areas.

[0019] Faced with the above problems, this application first considers establishing a dynamic correlation mechanism between plant spectral sensitivity and growth stage. In response to the problem of fixed spectral parameters in traditional methods, it attempts to introduce a collaborative analysis framework of plant variety data and environmental parameters. By quantifying the degree of photoperiod interference, it achieves accurate assessment. Further analysis reveals that simply relying on spectral intensity adjustment cannot solve the differentiated needs in multi-variety mixed planting scenarios. It is necessary to construct a spectral target curve by combining the characteristics of plant growth stages, obtain real-time growth status by fusing image recognition technology, and design a comprehensive spectral mismatch index as an optimization target, ultimately forming a closed-loop control logic.

[0020] Exemplary methods

[0021] Figure 1 The illustration depicts a computer vision-based landscape design method according to an embodiment of this application, comprising: receiving plant variety data, environmental parameter data, and nighttime spectral distribution data for each nighttime lighting area within a target garden; extracting corresponding spectral sensitivity functions based on the plant variety data and a preset plant spectral sensitivity function library; calculating a light impact factor for the degree of light interference of each nighttime lighting area on the photoperiod of plants within the area based on the environmental parameter data and the nighttime spectral distribution data; determining whether the light impact factor exceeds a preset impact threshold, and if so: acquiring image data of the nighttime lighting area and identifying the current growth stage of each plant variety within the area; extracting the target spectral curve of the corresponding plant variety from a preset spectral curve database based on the current growth stage; calculating a comprehensive spectral mismatch index within the nighttime lighting area based on the target spectral curve, the nighttime spectral distribution data, and the spectral sensitivity function; and, based on the comprehensive spectral mismatch index, solving and optimizing a target output spectrum for guiding the adjustment of nighttime lighting areas under the condition of satisfying preset visual brightness and rate of change constraints.

[0022] Among them, plant variety data refers to the information on plant species in each nighttime lighting area within the target garden. This information can be obtained through sensor collection or manual input and is used to subsequently match the spectral sensitivity function corresponding to the plant.

[0023] Among them, environmental parameter data refers to the values ​​of environmental factors such as temperature and humidity that affect plant growth. Specifically, these can be monitored in real time using weather stations or IoT devices to assess the regulatory effect of the environment on the photoperiod of plants affected by light.

[0024] Among them, nighttime spectral distribution data refers to the distribution information of light intensity at different wavelengths in the nighttime lighting area. Specifically, it can be collected using a spectrometer or a high-precision camera to quantify the actual impact of light on plants.

[0025] The spectral sensitivity function library refers to a set of functions pre-established for common plant varieties in the target garden, describing the response characteristics of each plant to different wavelengths of light. This function library contains spectral sensitivity data for each plant in relevant bands such as visible light and near-infrared light, reflecting the differences in the plant's ability to absorb, utilize, or respond to light of different wavelengths. Specific data can be obtained through experimental measurement, literature collection, or machine learning methods, and stored as standardized spectral sensitivity curves or function models.

[0026] Image data refers to visual information of the nighttime illuminated area, which can be collected using infrared cameras or multispectral imaging equipment to identify plant growth stages and spatial distribution.

[0027] The current growth stage refers to the specific developmental state of the plant in its life cycle. Specifically, it can be classified using image recognition algorithms combined with growth models to dynamically adjust the spectral target curve.

[0028] The spectral curve database refers to a set of standardized target spectral curves pre-established for different growth stages of various plant varieties in the target garden. This database records the ideal or suitable spectral distribution curves of each plant at each growth stage, reflecting the plant's requirements for light wavelength and intensity at different developmental states. The data sources of the spectral curve database include experimental measurements, literature data, and growth model predictions. It is stored in the form of digital curves or functions and covers multidimensional spectral information in the visible and near-infrared bands.

[0029] The target output spectrum refers to the optimal spectral output that minimizes the overall spectral mismatch while ensuring that the visual brightness and rate of change both meet the preset standards.

[0030] The core innovation of this application lies in dynamically monitoring plant growth stages and environmental parameters, and combining spectral sensitivity functions with multi-source data fusion calculations to enable nighttime landscape lighting to adaptively adjust spectral distribution while satisfying visual artistic effects, avoiding interference with plant photoperiods and ensuring the ecological health of the garden.

[0031] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0032] A nighttime lighting system will be installed in the cherry blossom viewing area of ​​a city park. The system will first receive data on plant species in the area, including cherry blossoms and ginkgo trees; environmental parameters such as temperature and humidity; and nighttime spectral distribution data of existing lighting equipment.

[0033] Spectral sensitivity functions for cherry blossoms and ginkgo at different growth stages are extracted from a pre-defined library of plant spectral sensitivity functions. Based on environmental parameters and spectral distribution data, a light impact factor is calculated. It is assumed that the calculated light impact factor exceeds a pre-defined threshold.

[0034] To further acquire high-resolution nighttime images of the area, image recognition algorithms were used to determine that cherry blossoms were in the flower bud differentiation stage and ginkgo trees were in the seedling stage. Spectral target curves for the cherry blossom flower bud differentiation stage and the ginkgo seedling stage were extracted from a spectral curve database.

[0035] Based on the extracted target spectral curve, actual spectral distribution data, and spectral sensitivity function, a comprehensive spectral mismatch index is calculated. Finally, while ensuring visual quality, a new target output spectrum is obtained through an optimization algorithm to guide the spectral adjustment of lighting equipment.

[0036] Through the above-described scheme, this application achieves precise control of garden nighttime lighting. By establishing a dynamic correlation mechanism between plant spectral sensitivity and growth stage, it solves the problem that traditional fixed spectral parameters cannot adapt to the diversity of plant varieties and changes in growth stages. An environmental parameter analysis framework is introduced to quantify the degree of photoperiodic interference, avoiding situations where light interference exceeds the plant's tolerance threshold. By combining real-time image recognition to obtain plant growth status, constructing a spectral target curve, and designing a comprehensive spectral mismatch index as an optimization target, a closed-loop control logic is formed. This method effectively reduces the interference of unreasonable lighting on normal plant growth and flowering, improves the ecological health of the garden, and simultaneously ensures the visual effect of the landscape.

[0037] In some of the schemes described above in this application, the calculation of the illumination influence factor specifically includes: using a set of discrete points based on spectral wavelengths. This represents the spectral wavelength sampling point of the nighttime illumination area i, where This represents the m-th wavelength sampling point; based on the nighttime spectral distribution data, construct the spectral distribution function values ​​of the nighttime illumination area i at each wavelength sampling point. Determine the set of plant species within the nighttime illumination area i based on plant species data. Where N is the number of plant varieties; the spectral sensitivity value at each wavelength sampling point is calculated based on the plant variety and its corresponding spectral sensitivity function. Based on spectral sensitivity values With spectral distribution function value Calculate each plant variety Effective illuminance of the spectrum in nighttime lighting area i For each plant variety Assign weights that reflect the importance of its sensitivity to photoperiod. ; Calculate the number of plant species within the nighttime lighting area i based on environmental parameter data. Environmental impact factors Environmental impact factors This indicates the degree to which the environment affects the photoperiod of plants due to nighttime light disturbance; based on the effective light intensity. Environmental impact factors and weighting coefficients Calculate the illumination influence factor within nighttime lighting area i. : .

[0038] Specifically, the spectral distribution function value is constructed by weighted averaging of the spectral response values ​​of multiple pixels, with the pixel signal-to-noise ratio serving as the weighting factor to reduce the impact of low-quality pixels on the overall data. The calculation of effective illuminance involves integrating the spectral sensitivity function, spectral distribution function, and light attenuation coefficient at discrete wavelengths, and then summing the results by multiplying the wavelength intervals and the attenuation coefficients to obtain the total effective illuminance. The environmental impact factor is calculated using a piecewise function. When the ambient temperature and humidity deviate from the suitable range for plants, linear reduction or zeroing is applied to reflect the aggravating effect of environmental degradation on photoperiodic interference. Finally, the illuminance impact factor is integrated by weighted summation of the sensitivity weights of each plant variety, effective light intake, and environmental impact factors to form a comprehensive quantitative index. This method improves the accuracy of illuminance interference assessment through spatial distance attenuation compensation and dynamic environmental correction, providing a reliable basis for subsequent lighting adjustments.

[0039] Through the above-described technical solution, this application can accurately calculate the impact of nighttime lighting on garden plants. By considering multiple aspects such as spectral distribution, plant sensitivity, and environmental factors, the impact of light on plant growth is comprehensively assessed. This quantitative assessment method can provide a scientific basis for garden landscape design, help optimize nighttime lighting schemes, reduce adverse effects on plant growth, and thus improve the ecological benefits of garden landscapes.

[0040] In some of the above-mentioned schemes of this application, the construction of the spectral distribution function value specifically includes: obtaining the original first spectral response value of each sampled pixel in the nighttime illumination area at each wavelength sampling point based on the nighttime spectral distribution data; performing radiation calibration and noise filtering on the spectral response value to obtain the second spectral response value that eliminates the influence of environmental interference and sensor error; and performing a weighted average of the second spectral response values ​​of all sampled pixels in the nighttime illumination area at each wavelength sampling point based on the pixel signal-to-noise ratio of each sampled pixel to obtain the spectral distribution function value of the nighttime illumination area at each wavelength sampling point.

[0041] The radiometric calibration process converts the original first spectral response value into standard radiometric units using a pre-defined sensor response model, eliminating response differences between different sensors. Noise filtering employs wavelet transform to separate high-frequency noise components while retaining the effective spectral signal. In the weighted averaging process, the pixel signal-to-noise ratio (SNR) is determined by calculating the ratio of the signal variance to the noise variance of each pixel across multiple consecutive frames. Pixels with higher SNR are given greater weight during averaging, effectively suppressing the influence of random noise on the spectral distribution function value.

[0042] Specifically, after obtaining the original first spectral response value, the voltage signal output by the sensor is converted into a radiance value through radiometric calibration, eliminating measurement deviations caused by sensor aging or temperature changes. Noise filtering further removes high-frequency noise components introduced by environmental electromagnetic interference or circuit noise. For example, the spectral response value is decomposed into five levels using a Daubechies wavelet basis, thresholded, and then reconstructed to obtain the second spectral response value. In the weighted averaging stage, the final spectral distribution function value of each wavelength sampling point is obtained by multiplying the second spectral response values ​​of all pixels at that wavelength by their normalized signal-to-noise ratio weights and then summing them. Pixels with high signal-to-noise ratios contribute more data, while the influence of pixels with low signal-to-noise ratios is weakened, thereby improving the overall accuracy of the spectral distribution function value. Through the above processing steps, the cumulative effects of environmental interference and sensor errors are effectively eliminated, ensuring the reliability of subsequent calculations of the degree of optical periodicity interference.

[0043] Through the above technical solutions, this application can effectively eliminate the influence of environmental interference and sensor errors on the spectral distribution function values, thereby improving the accuracy of the spectral distribution function values. Furthermore, by weighted averaging of multiple sampled pixels, the random error of individual pixels can be reduced, improving the stability and representativeness of the spectral distribution function values. This provides a more reliable data foundation for subsequent landscape design based on spectral distribution, and helps to achieve more precise lighting control.

[0044] In some of the schemes described above in this application, the effective illuminance is calculated. Specifically, this includes: extracting plant species based on image data of nighttime illuminated area i. Spatial distance from the light source Based on spatial distance Calculation and Plant Varieties Corresponding light attenuation coefficient :

[0045] ;

[0046] in, To avoid using tiny positive constants with a denominator of zero;

[0047] Calculate effective illuminance :

[0048] ;

[0049] in, The interval between adjacent wavelength sampling points.

[0050] Specifically, the light attenuation coefficient is used to simulate the physical law of light intensity decreasing with the square of distance, while the small constant ε is used to handle the extreme case where the distance is zero. In the calculation of effective illuminance, the spectral sensitivity value, the spectral distribution function value, and the attenuation coefficient are multiplied point-by-point at discrete wavelengths, and the sums are multiplied by the wavelength interval to achieve an integral approximation, ultimately obtaining the effective illuminance received by the plant variety at a specific spatial location. This process improves the accuracy of effective illuminance calculation by quantifying the influence of distance on light intensity, providing a data foundation for the accurate assessment of subsequent photoperiodic interference levels.

[0051] Through the above technical solution, this application can accurately calculate the effective light intensity received by plants. This allows for a more precise assessment of the impact of nighttime lighting on plant growth, avoiding the adverse effects of excessive lighting. Furthermore, this method considers the spatial distance between the plant and the light source, making the light intensity calculation more realistic and providing a scientific basis for nighttime lighting design in landscape architecture.

[0052] In some of the above-mentioned solutions in this application, the extraction of spatial distance specifically involves: performing image segmentation on the image data of the nighttime lighting area to obtain the pixel regions of each plant species in the image data; converting the pixel regions of each plant species into a corresponding set of three-dimensional spatial coordinate points based on the calibration information of the image data, with the origin of the three-dimensional space being the position of the light source in the nighttime lighting area; calculating the centroid of the set of three-dimensional spatial coordinate points corresponding to each plant species as the spatial coordinates of that plant species; and calculating the Euclidean distance between the spatial coordinates of each plant species and the position of the light source as the spatial distance.

[0053] Image segmentation employs deep learning-based semantic segmentation algorithms, such as U-Net or Mask R-CNN, to distinguish pixel regions of different plant species. Calibration information includes the camera intrinsic matrix and the extrinsic matrix of the light source position, converting pixel coordinates into three-dimensional spatial coordinates using a perspective projection model. Centroid calculation is achieved by taking the geometric center of the set of three-dimensional coordinate points, for example, by taking the average value of each coordinate point along the x, y, and z axes. The combination of image segmentation and calibration information ensures the accuracy of plant region localization, while the three-dimensional coordinate transformation and centroid calculation eliminate the influence of uneven pixel distribution, thereby improving the accuracy of spatial distance measurement.

[0054] Specifically, image segmentation algorithms are used to separate pixel regions of different plant species from images of nighttime illuminated areas, avoiding background interference. Using pre-calibrated camera parameters and light source positions, the segmented pixel regions are mapped onto a 3D coordinate system with the light source as the origin, ensuring spatial consistency. The centroid of the 3D coordinate set for each plant species is calculated to eliminate coordinate deviations caused by irregular plant morphology or partial occlusion. Finally, based on the Euclidean distance between the centroid coordinates and the light source, consistent and reliable spatial distance data is provided for each plant species. For example, when the pixel regions of a plant species are discrete due to uneven leaf distribution, centroid calculation can effectively smooth local fluctuations, making distance measurements closer to the actual physical location. This process, through precise spatial positioning and distance calculation, provides reliable input for the subsequent determination of the light attenuation coefficient, thereby optimizing the evaluation results of effective light intensity.

[0055] In some of the schemes described above in this application, environmental impact factors are calculated. Specifically, it includes:

[0056] The average temperature data of nighttime lighting area i within a preset time period is extracted based on environmental parameter data. and average humidity data ;

[0057] Based on average temperature data Calculate the temperature influence factor :

[0058] ;

[0059] in, The optimal temperature for the current plant varieties. This refers to the permissible temperature deviation range for the current plant variety. The preset temperature weighting coefficient;

[0060] Based on humidity data Calculate the influence factor of humidity :

[0061] ;

[0062] in, The optimal humidity for the current plant varieties, This represents the permissible humidity deviation range for the current plant variety. This is the preset humidity weighting coefficient;

[0063] Environmental impact factors are calculated based on temperature and humidity influence factors. : ,in, and The preset weighting coefficients satisfy... .

[0064] In other words, the temperature influence factor is calculated using a piecewise function. When the difference between the regional temperature and the optimal temperature for plants is within the allowable deviation range, the temperature influence factor decreases linearly with increasing temperature difference; otherwise, it returns to zero. For every 1°C increase in the absolute value of the temperature difference, the temperature influence factor decreases by a factor of α. The humidity influence factor is calculated using the same logic: when the humidity difference is within the allowable deviation range, the humidity influence factor decreases linearly with increasing humidity difference; otherwise, it returns to zero. For every 1% increase in the absolute value of the humidity difference, the humidity influence factor decreases by a factor of β. The temperature and humidity influence factors are weighted and summed using preset weighting coefficients, with the total sum of the weighting coefficients being 1.

[0065] Specifically, taking a certain nighttime lighting area as an example, the optimal temperature for the current plant species is preset to 25℃, with an allowable temperature deviation range of ±5℃, and the temperature weighting coefficient α is set to 0.02. When the average temperature of the area is 23℃, the absolute temperature difference is 2℃, which is within the allowable range, and the temperature influence factor is calculated as 1 - 0.02 × 2 = 0.96. If the average humidity is 65%, the optimal humidity is 70%, the allowable humidity deviation range is ±10%, and the humidity weighting coefficient β is set to 0.01, then the absolute humidity difference is 5%, and the humidity influence factor is calculated as 1 - 0.01 × 5 = 0.95. Assuming the temperature weighting coefficient w_T is 0.6 and the humidity weighting coefficient w_H is 0.4, the final environmental influence factor is 0.6 × 0.96 + 0.4 × 0.95 = 0.956. This value indicates that the current environmental conditions are within a controllable range of interference with plant photoperiod, providing accurate correction parameters for subsequent calculations of the light influence factor. By introducing a segmented calculation model for temperature and humidity, the impact of environmental parameter fluctuations on plant photoperiod can be dynamically reflected, avoiding biases caused by assessments of single environmental factors.

[0066] Through the above technical solution, this application can more accurately assess the impact of environmental factors on plant photoperiod disturbance. By considering the two key environmental parameters of temperature and humidity and introducing a weighting system, the calculation of environmental impact factors becomes more comprehensive and accurate. This helps to better balance nighttime lighting effects and plant growth needs in landscape design, thereby improving the overall ecological health and aesthetics of the garden.

[0067] In some of the schemes described above in this application, the calculation of the comprehensive spectral mismatch index specifically includes: obtaining the sampling points of the nighttime illumination region i at each wavelength. Current spectral distribution value For each plant species within the nighttime illumination area i The target spectral value corresponding to its current growth stage is obtained based on the target spectral curve. ; Calculate each plant variety Single-variety spectrum mismatch :

[0068] ;

[0069] in, Plant varieties At wavelength sampling point Corresponding spectral sensitivity;

[0070] According to the weighting coefficient Calculate the comprehensive spectral mismatch index within illumination region i. :

[0071] .

[0072] In the above scheme, the spectral mismatch degree of a single variety is calculated using a weighted average difference, with the weight representing the spectral sensitivity value of the plant variety at the corresponding wavelength. For example, a plant with high sensitivity at 500 nm will have a greater contribution to the mismatch degree due to its spectral deviation at that wavelength. The comprehensive index is calculated by linearly superimposing the weight coefficients of each variety. The weight coefficients reflect the importance of the plant's sensitivity to photoperiod; for example, flowering plants are given higher weight coefficients. During the calculation, the wavelength sampling interval is set to 5 nm to ensure that the spectral resolution meets the requirements for analyzing the plant's photosensitivity characteristics.

[0073] Specifically, the spectral distribution values ​​of the nighttime illumination area are collected using a multispectral imaging device, and the light intensity data at each wavelength is obtained after radiometric calibration. Target spectral values ​​are extracted from a preset database, such as the spectral curve of a specific peak in the 380-700 nm range for ginkgo trees in their vegetative growth stage. When calculating the spectral mismatch degree for a single variety, if the current spectral value is detected to be 15% higher than the target value at a wavelength of 680 nm, this deviation is amplified by a sensitivity weight of 0.8 and included in the total deviation. In the calculation of the comprehensive index, the weighting coefficient for flowering peonies is set to 0.6, and for dormant shrubs, it is set to 0.2. The final weighted result reflects the overall degree of spectral interference on different plants. When this index is used to optimize the light spectrum, priority is given to adjusting the light intensity of wavelengths that significantly affect high-weight plant varieties; for example, reducing the intensity of red light at 650 nm by 20% to reduce photoperiodic interference on flowering plants.

[0074] Through the aforementioned technical solution, this application can accurately quantify the impact of nighttime lighting on garden plants. By calculating the comprehensive spectral mismatch index, the difference between the current lighting scheme and plant needs can be objectively assessed, providing a scientific basis for subsequent lighting adjustments. This data-driven approach avoids the uncertainty of subjective experience-based judgments, helps optimize nighttime lighting design, and minimizes interference with plant growth while ensuring landscape effects. Furthermore, by considering the spectral sensitivity and importance of different plant species, this method can more comprehensively balance the needs of various plants and improve the overall health of the garden ecosystem.

[0075] Exemplary System

[0076] Figure 2 The illustration shows a computer vision-based landscape design system according to an embodiment of this application, comprising: a data acquisition module for receiving plant variety data, environmental parameter data, and nighttime spectral distribution data for each nighttime lighting area within a target garden; a function extraction module for extracting corresponding spectral sensitivity functions based on plant variety data and a preset plant spectral sensitivity function library; a first calculation module for calculating the light impact factor of each nighttime lighting area on the degree of photoperiod interference of the light on plants within the area based on environmental parameter data and nighttime spectral distribution data; a judgment module for judging whether the light impact factor exceeds a preset impact threshold; an image recognition module for acquiring image data of the nighttime lighting area and identifying the current growth stage of each plant variety within the area; a target curve extraction module for extracting the spectral target curve of the corresponding plant variety from a preset spectral curve database based on the current growth stage; a second calculation module for calculating a comprehensive spectral mismatch index within the nighttime lighting area based on the spectral target curve, nighttime spectral distribution data, and spectral sensitivity functions; and a third calculation module for solving and optimizing the target output spectrum for guiding the adjustment of nighttime lighting areas based on the comprehensive spectral mismatch index, under the condition of satisfying preset visual brightness and rate of change constraints.

[0077] In one example, the first calculation module calculates the illumination influence factor by: a set of discrete points based on spectral wavelengths. This represents the spectral wavelength sampling point of the nighttime illumination area i, where This represents the m-th wavelength sampling point; based on the nighttime spectral distribution data, construct the spectral distribution function values ​​of the nighttime illumination area i at each wavelength sampling point. Determine the set of plant species within the nighttime illumination area i based on plant species data. Where N is the number of plant varieties; the spectral sensitivity value at each wavelength sampling point is calculated based on the plant variety and its corresponding spectral sensitivity function. Based on spectral sensitivity values With spectral distribution function value Calculate each plant variety Effective illuminance of the spectrum in nighttime lighting area i For each plant variety Assign weights that reflect the importance of its sensitivity to photoperiod. ; Calculate the number of plant species within the nighttime lighting area i based on environmental parameter data. Environmental impact factors Environmental impact factors This indicates the degree to which the environment affects the photoperiod of plants due to nighttime light disturbance; based on the effective light intensity. Environmental impact factors and weighting coefficients Calculate the illumination influence factor within nighttime lighting area i. : ...

[0078] In one example, the first calculation module constructs the spectral distribution function value by: obtaining the original first spectral response value of each sampled pixel in the nighttime illumination area at each wavelength sampling point based on the nighttime spectral distribution data; performing radiometric calibration and noise filtering on the spectral response value to obtain a second spectral response value that eliminates the influence of environmental interference and sensor error; and performing a weighted average of the second spectral response values ​​of all sampled pixels in the nighttime illumination area at each wavelength sampling point based on the pixel signal-to-noise ratio of each sampled pixel to obtain the spectral distribution function value of the nighttime illumination area at each wavelength sampling point.

[0079] In one example, the first calculation module calculates the effective illuminance. Includes: extracting plant species based on image data of nighttime illuminated area i Spatial distance from the light source Based on spatial distance Calculation and Plant Varieties Corresponding light attenuation coefficient :

[0080] ;

[0081] in, To avoid using tiny positive constants with a denominator of zero;

[0082] Calculate effective illuminance :

[0083] ;

[0084] in, The interval between adjacent wavelength sampling points.

[0085] In one example, the first calculation module extracts spatial distance by: performing image segmentation on the image data of the nighttime lighting area to obtain the pixel regions of each plant species in the image data; converting the pixel regions of each plant species into a corresponding set of three-dimensional spatial coordinate points based on the calibration information of the image data, with the origin of the three-dimensional space being the position of the light source in the nighttime lighting area; calculating the centroid of the set of three-dimensional spatial coordinate points corresponding to each plant species as the spatial coordinates of that plant species; and calculating the Euclidean distance between the spatial coordinates of each plant species and the position of the light source as the spatial distance.

[0086] In one example, the first calculation module calculates the environmental impact factor. This includes: extracting the average temperature data of nighttime lighting area i within a preset time period based on environmental parameter data. and average humidity data Based on average temperature data Calculate the temperature influence factor :

[0087] ;

[0088] in, The optimal temperature for the current plant varieties. This refers to the permissible temperature deviation range for the current plant variety. The preset temperature weighting coefficient;

[0089] Based on humidity data Calculate the influence factor of humidity :

[0090] ;

[0091] in, The optimal humidity for the current plant varieties, This represents the permissible humidity deviation range for the current plant variety. The humidity weighting coefficient is preset; the environmental impact factor is calculated based on the temperature and humidity influence factors. : ,in, and The preset weighting coefficients satisfy... .

[0092] In one example, the second calculation module calculates the overall spectral mismatch index by: obtaining the sampling points of the nighttime illumination area i at each wavelength. Current spectral distribution value For each plant species within the nighttime illumination area i The target spectral value corresponding to its current growth stage is obtained based on the target spectral curve. ; Calculate each plant variety Single-variety spectrum mismatch :

[0093] ;

[0094] in, Plant varieties At wavelength sampling point Corresponding spectral sensitivity;

[0095] According to the weighting coefficient Calculate the comprehensive spectral mismatch index within illumination region i. :

[0096] .

[0097] Exemplary electronic devices

[0098] Figure 3 An electronic device according to an embodiment of this application is illustrated. The electronic device may be the mobile device itself, or a standalone device independent of it, which may communicate with the mobile device to receive collected input signals from it and send selected target driving behaviors to it.

[0099] Figure 3 A block diagram of an electronic device according to an embodiment of this application is illustrated.

[0100] like Figure 3 As shown, the electronic device includes one or more processors and memory.

[0101] A processor can be a central processing unit (CPU) or other form of processing unit with data processing and / or instruction execution capabilities, and can control other components in an electronic device to perform desired functions.

[0102] The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and a processor may execute the program instructions to implement the driving behavior decision-making methods of the various embodiments of this application described above, and / or other desired functions.

[0103] In one example, the electronic device may also include input devices and output devices, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0104] Of course, for the sake of simplicity, Figure 3 Only some of the components of the electronic device relevant to this application are shown in this illustration; components such as buses, input / output interfaces, etc., are omitted. In addition, the electronic device may include any other suitable components depending on the specific application.

[0105] Exemplary computer-readable media

[0106] Embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the driving behavior decision-making methods according to various embodiments of this application described in the "Exemplary Methods" section above.

[0107] Computer-readable storage media may take the form of any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0108] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0109] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0110] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0111] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0112] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A computer vision-based garden landscape design method, characterized by, include: Receive plant species data, environmental parameter data, and nighttime spectral distribution data for each nighttime lighting area within the target garden; Extract the corresponding spectral sensitivity function based on the plant variety data and the preset plant spectral sensitivity function library; Based on the environmental parameter data and nighttime spectral distribution data, calculate the light influence factor of the light on the degree of interference of light on the photoperiod of plants in each nighttime lighting area; Determine whether the illumination influence factor exceeds a preset influence threshold; if so, then: Acquire image data of the nighttime illuminated area and identify the current growth stage of each plant species within the area; Based on the current growth stage, extract the target spectral curve of the corresponding plant variety from the preset spectral curve database; Based on the target spectral curve, the nighttime spectral distribution data, and the spectral sensitivity function, calculate the comprehensive spectral mismatch index within the nighttime illumination area; Based on the comprehensive spectral mismatch index, and under the condition of satisfying the preset visual brightness and rate of change constraints, the target output spectrum used to guide the adjustment of the nighttime lighting area is obtained by solving the optimization. The calculation of the illumination influence factor includes the following steps: The set of discrete points based on spectral wavelength The sampling point for the spectral wavelength of the nighttime illumination area i is represented, where This represents the m-th wavelength sampling point; constructing a spectral distribution function value of the night lighting area i at each wavelength sampling point according to the night spectral distribution data ; determining a set of plant varieties within the night-time lighting zone i from the plant variety data where N is the number of plant varieties. According to the plant variety and the corresponding spectral sensitivity function, the spectral sensitivity value of each wavelength sampling point is calculated ; According to the spectral sensitivity values With the spectral distribution function values Calculating for each plant variety The effective light exposure of the plant in the night lighting area i to the spectral illumination ; for each plant variety assigning a weight coefficient reflecting the importance of its sensitivity to photoperiod ; Calculate the number of plant species within the nighttime lighting area i based on the environmental parameter data. Environmental impact factors The environmental impact factors This indicates the degree to which the environment affects the photoperiod of plants due to nighttime light disturbance; According to the effective illuminance Environmental impact factors and the weighting coefficients Calculate the illumination influence factor within the nighttime illumination area i. : . 2.The computer vision based landscape design method of claim 1, wherein, The constructed spectral distribution function values ​​include: Based on the nighttime spectral distribution data, obtain the original first spectral response value of each sampled pixel in the nighttime illumination area at each wavelength sampling point; The spectral response value is subjected to radiometric calibration and noise filtering to obtain a second spectral response value that eliminates the influence of environmental interference and sensor error; The second spectral response values ​​of all sampled pixels within the nighttime illumination area are weighted and averaged at each wavelength sampling point according to the pixel signal-to-noise ratio of each sampled pixel to obtain the spectral distribution function value of the nighttime illumination area at each wavelength sampling point.

3. The computer vision-based landscape design method according to claim 1, characterized in that, The calculated effective light exposure comprising: Extract the plant species based on the image data of the nighttime illumination area i. Spatial distance from the light source ; According to the spatial distance Computing the plant variety Corresponding light attenuation coefficient : ； wherein To avoid a tiny positive constant for the denominator being zero; calculating the effective light exposure : ； wherein is the interval of adjacent wavelength sampling points.

4. The computer vision-based landscape design method according to claim 3, characterized in that, The extraction of spatial distance specifically refers to: Image segmentation is performed on the image data of the nighttime illuminated area to obtain the pixel regions of each plant species in the image data; Based on the calibration information of the image data, the pixel regions of each plant variety are converted into a corresponding set of three-dimensional spatial coordinate points, and the origin of the three-dimensional space is the position of the light source in the nighttime lighting area. Calculate the centroid of the set of three-dimensional spatial coordinate points corresponding to each plant variety, and use it as the spatial coordinate of that plant variety; The spatial distance is calculated based on the Euclidean distance between the plant species and the location of the light source.

5. The computer vision-based landscape design method according to claim 1, characterized in that, The computing environment impact factor comprises: extracting average temperature data of the night lighting area i in a preset time period according to the environmental parameter data and average humidity data ; According to the average temperature data Computing temperature impact factors : ； wherein, an optimum temperature suitable for the current plant variety, a temperature deviation range allowed for the current plant variety, a preset temperature weight coefficient; According to the humidity data Computing a humidity impact factor : ； wherein, an optimal humidity suitable for the current plant variety, a humidity deviation range allowed for the current plant variety, a preset humidity weight coefficient; The environmental impact factor is calculated based on the temperature impact factor and the humidity impact factor. : ,in, and The preset weighting coefficients satisfy... .

6. The computer vision-based landscape design method according to claim 1, characterized in that, The index for calculating the overall spectrum mismatch includes: Obtain the nighttime illumination area i at each wavelength sampling point Current spectral distribution value ; For each plant variety within the night lighting area i , according to the light spectrum target curve, the target light spectrum value corresponding to its current growth stage is obtained ; Calculating a single variety profile mismatch for each plant variety :​ ； wherein is a plant variety at wavelength sampling points corresponding spectral sensitivity; According to the weighting coefficients Calculate the comprehensive spectral mismatch index within the illumination region i. : 。 7. A computer vision based landscape design system characterized in that, include: The data acquisition module is used to receive plant variety data, environmental parameter data, and nighttime spectral distribution data for each nighttime lighting area within the target garden. The function extraction module is used to extract the corresponding spectral sensitivity function based on the plant variety data and a preset plant spectral sensitivity function library. The first calculation module is used to calculate the light influence factor of the light on the degree of interference of light on the photoperiod of plants in each nighttime lighting area based on the environmental parameter data and the nighttime spectral distribution data. The judgment module is used to determine whether the illumination influence factor exceeds a preset influence threshold. An image recognition module is used to acquire image data of the nighttime illumination area and identify the current growth stage of each plant species within the area; The target curve extraction module is used to extract the target spectral curve of the corresponding plant variety from a preset spectral curve database according to the current growth stage. The second calculation module is used to calculate the comprehensive spectral mismatch index in the nighttime illumination area based on the spectral target curve, the nighttime spectral distribution data, and the spectral sensitivity function. The third calculation module is used to solve and optimize the target output spectrum for guiding the adjustment of the nighttime lighting area based on the comprehensive spectrum mismatch index, under the condition of satisfying the preset visual brightness and rate of change constraints. The first calculation module calculates the illumination impact factor by including the following steps: The set of discrete points based on spectral wavelength The sampling point for the spectral wavelength of the nighttime illumination area i is represented, where This represents the m-th wavelength sampling point; Construct the spectral distribution function values ​​of the nighttime illumination area i at each wavelength sampling point based on the nighttime spectral distribution data. ; The set of plant species within the nighttime lighting area i is determined based on the plant species data. , where N is the number of plant varieties; The spectral sensitivity values ​​at each wavelength sampling point were calculated based on the plant variety and its corresponding spectral sensitivity function. ; Based on spectral sensitivity value With the spectral distribution function value Calculate each plant variety Effective illuminance of the spectrum irradiated in the nighttime illumination area i ; For each plant variety Assign weights that reflect the importance of its sensitivity to photoperiod. ; Calculate the number of plant species within the nighttime lighting area i based on the environmental parameter data. Environmental impact factors The environmental impact factors This indicates the degree to which the environment affects the photoperiod of plants due to nighttime light disturbance; According to the effective illuminance Environmental impact factors and the weighting coefficients Calculate the illumination influence factor within the nighttime illumination area i. : .

8. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the method as described in any one of claims 1 to 6.

9. A computer storage medium storing computer-executable instructions thereon, characterized in that: When the computer-executable instructions are executed by a processor, they implement the steps of the method as described in any one of claims 1 to 6.

Images