Landscape landscaping environment monitoring device

Abstract

The invention provides a landscape garden greening environment monitoring device, and relates to the technical field of environment monitoring, and the device comprises an obtaining module which is used for obtaining image data of a landscape garden; the extraction module is used for extracting a plurality of vegetation features from the image data; the screening module is used for screening the plurality of vegetation features and determining three vegetation features with the strongest relevance with the vegetation health degree; the standardization module is used for standardizing characteristic values of the three vegetation characteristics to be between 0 and 255; the fusion module is used for fusing the first vegetation feature as an R channel, the second vegetation feature as a G channel and the third vegetation feature as a B channel to form a color vegetation feature map; and the monitoring module is used for monitoring the health state of the vegetation in the landscape garden through a convolutional neural network by taking the color vegetation feature map as input. According to the invention, the monitoring efficiency of landscaping can be improved, accurate and real-time health assessment is provided, the labor cost is reduced, and intelligent garden management is realized.

Description

Technical Field

[0001] The invention relates to the technical field of environmental monitoring, and in particular to a landscape gardening environment monitoring device. Background Art

[0002] By monitoring the health of landscape gardens, we can ensure that plants grow well and avoid yellowing leaves and withering caused by pests and diseases, climate change, etc., thereby maintaining the beauty of the gardens. At the same time, plants in landscape gardens play an important role in absorbing carbon dioxide, releasing oxygen, and regulating temperature and humidity. By monitoring the health of landscape greening, we can promptly discover and solve vegetation health problems and prevent the destruction of the garden ecosystem.

[0003] At present, the monitoring of landscape gardening and greening environments is still mainly based on manual inspections. Landscape and greening managers conduct regular inspections, mainly through visual inspections of vegetation health, such as leaf color changes, wilting, and signs of pests and diseases.

[0004] However, manual inspections require managers to check the vegetation in each area one by one, especially in large-scale gardens. This method is time-consuming and laborious, and it is difficult to achieve comprehensive monitoring of all areas. At the same time, due to limited personnel resources, inspections are usually carried out on a regular basis, making it difficult to achieve continuous real-time monitoring. This means that some unexpected problems (such as pests and diseases, drought, etc.) may not be discovered and handled in time. Large-scale gardens require a large amount of manpower for inspections and maintenance. The high labor cost of manual inspections is difficult to bear in long-term management, especially in the case of a shortage of human resources, which may lead to insufficient or untimely inspections. Summary of the invention

[0005] In order to solve the technical problems that manual inspections are difficult to achieve comprehensive monitoring of all areas, cannot promptly discover and handle sudden problems, and have high labor costs, the present invention provides a landscape gardening environment monitoring device.

[0006] The technical solution provided by the embodiment of the present invention is as follows:

[0007] A landscape gardening environment monitoring device provided by an embodiment of the present invention includes: an acquisition module for acquiring image data of a landscape garden;

[0008] An extraction module, used for extracting a plurality of vegetation features from the image data;

[0009] A screening module is used to screen multiple vegetation features and determine the three vegetation features that are most closely related to vegetation health;

[0010] The normalization module is used to normalize the characteristic values ​​of the three vegetation features to between 0 and 255;

[0011] A fusion module, used to fuse the first vegetation feature as an R channel, the second vegetation feature as a G channel, and the third vegetation feature as a B channel to form a color vegetation feature map;

[0012] The monitoring module is used to monitor the health status of vegetation in the landscape garden by using the color vegetation feature map as input through a convolutional neural network.

[0013] Furthermore, the extraction module is specifically used for:

[0014] Extract the green sum index SGI of each region from the conventional image;

[0015] The enhanced vegetation index EVI, moisture stress index MSI, structure-insensitive pigment index SIPI, atmospheric resistance vegetation index ARVI, normalized difference vegetation index NDVI, green normalized difference vegetation index GNDVI and blue normalized difference vegetation index BNDVI of each area are extracted from infrared remote sensing images.

[0016] Furthermore, the screening module is specifically used for:

[0017] Linear fitting is performed between multiple vegetation characteristics and vegetation health to determine the regression coefficient of each vegetation characteristic;

[0018] The vegetation characteristics were sorted in descending order according to the regression coefficient, and the top three vegetation characteristics were selected as the vegetation characteristics with the strongest correlation with vegetation health.

[0019] The normalization module is specifically used for:

[0020] Linear normalization is used to normalize the characteristic values ​​of the three vegetation features to between 0 and 255.

[0021] The monitoring module is specifically used for:

[0022] The input layer, the first convolution unit, the second convolution unit, and the third convolution unit are sequentially connected, and the outputs of each convolution unit are cascaded to construct a convolutional neural network with a cascade structure;

[0023] Inputting the color vegetation feature map into the input layer;

[0024] Extracting a first feature map through the first convolution unit, extracting a second feature map through the second convolution unit, and extracting a third feature map through the third convolution unit;

[0025] The third feature map is used to monitor the health status of vegetation alone to obtain a first monitoring result, the second feature map and the third feature map are cascaded and fused to monitor the health status of vegetation to obtain a second monitoring result, and the first feature map, the second feature map and the third feature map are cascaded and fused to monitor the health status of vegetation to obtain a third monitoring result;

[0026] The first monitoring result, the second monitoring result and the third monitoring result are integrated to determine a final monitoring result.

[0027] The health state with the largest probability value is taken as the final monitoring result of this health state monitoring.

[0028] Furthermore, the health status includes: health, mild stress, moderate stress, severe stress and dying.

[0029] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0030] In the present invention, through the landscape gardening environment monitoring device, a convolutional neural network is used to automatically monitor the health status of vegetation in all areas of the landscape garden, which can improve the monitoring efficiency of landscape gardening, provide more accurate and real-time health assessments, reduce labor costs and subjective errors, and realize intelligent and scientific garden management. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0032] Figure 1 A schematic diagram of the structure of a landscape gardening environment monitoring device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0034] Reference Manual Attached Figure 1 , showing a schematic structural diagram of a landscape gardening environment monitoring device provided by an embodiment of the present invention.

[0035] The embodiment of the present invention provides a landscape gardening environment monitoring device, which is characterized by comprising:

[0036] The acquisition module 1 is used to acquire image data of landscape architecture.

[0037] Among them, image data may include conventional images, infrared remote sensing images, depth images, etc.

[0038] The extraction module 2 is used to extract multiple vegetation features from the image data.

[0039] In a possible implementation, the extraction module 2 is specifically used to extract the green sum index SGI of each area from the conventional image. Extract the enhanced vegetation index EVI, moisture stress index MSI, structural insensitive pigment index SIPI, atmospheric resistance vegetation index ARVI, normalized difference vegetation index NDVI, green normalized difference vegetation index GNDVI, and blue normalized difference vegetation index BNDVI of each area from the infrared remote sensing image.

[0040] Furthermore, the green sum index is as follows:

[0041]

[0042] Where SGI represents the green sum index, ρ i It represents the reflectance in the green band range of 500-600nm, and N represents the number of bands in the green band.

[0043] Furthermore, the enhanced vegetation index is specifically:

[0044]

[0045] Among them, EVI stands for Enhanced Vegetation Index, NIR stands for the reflectivity of the near-infrared band, RED stands for the reflectivity of the red light band, and BLUE stands for the reflectivity of the blue light band.

[0046] Furthermore, the humidity stress index is specifically:

[0047]

[0048] Where MSI represents the moisture stress index, ρ 1599 Represents the reflectivity of the 1599nm band, ρ 819 Indicates the reflectivity in the 819nm band.

[0049] Furthermore, the structure-insensitive pigment index is specifically:

[0050]

[0051] Where SIPI stands for structure-insensitive pigment index, ρ 800 Represents the reflectivity in the 800nm ​​band, ρ 445 Indicates the reflectivity in the 445nm band, ρ 680Indicates the reflectivity in the 680nm band.

[0052] Furthermore, the atmospheric resistance vegetation index is specifically:

[0053]

[0054] Among them, ARVI represents the atmospheric resistance vegetation index, NIR represents the reflectivity of the near-infrared band, RED represents the reflectivity of the red light band, BLUE represents the reflectivity of the blue light band, and γ represents the adjustment factor used to correct the atmospheric effect.

[0055] Furthermore, the normalized difference vegetation index is specifically:

[0056]

[0057] Among them, NDVI represents the normalized difference vegetation index, NIR represents the reflectivity of the near-infrared band, and RED represents the reflectivity of the red band.

[0058] Furthermore, the green normalized difference vegetation index is specifically:

[0059]

[0060] Among them, NDVI represents the normalized difference vegetation index, NIR represents the reflectance of the near-infrared band, and GREEN represents the reflectance of the green band.

[0061] Furthermore, the blue normalized difference vegetation index is specifically:

[0062]

[0063] Among them, BNDVI represents the normalized difference vegetation index, NIR represents the reflectance of the near-infrared band, and BLUE represents the reflectance of the blue light band.

[0064] Screening module 3 is used to screen multiple vegetation features and determine the three vegetation features that are most closely related to vegetation health.

[0065] In a possible implementation, the screening module 3 is specifically used to screen multiple vegetation features through principal component analysis to determine the three vegetation features that are most strongly correlated with vegetation health.

[0066] Among them, principal component analysis (PCA) is a widely used dimensionality reduction technology, which is mainly used to extract the most important information from high-dimensional data, project the data into a low-dimensional space, reduce the data dimension, and try to retain the main features and variance information of the data.

[0067] In the present invention, during the landscape monitoring process, many vegetation features may be extracted, some of which have strong correlations and carry redundant information. PCA eliminates redundant information by analyzing the linear correlations between features, and retains the features with the strongest correlation with vegetation health.

[0068] In a possible implementation, the screening module 3 is specifically used to screen multiple vegetation features through a linear regression analysis method to determine the three vegetation features that are most strongly correlated with vegetation health.

[0069] Furthermore, the screening module 3 is specifically used for:

[0070] Linear fitting is performed on multiple vegetation characteristics and vegetation health to determine the regression coefficient of each vegetation characteristic:

[0071] τ=β0+β1a1+β2a2+L+β n a n

[0072] Among them, τ represents the vegetation health, a i represents the i-th vegetation feature, β i represents the regression coefficient of the ith vegetation feature, and β0 represents the constant term.

[0073] It should be noted that the size of the regression coefficient reflects the influence of the corresponding feature on vegetation health, so that the most important features can be accurately selected for subsequent analysis.

[0074] The vegetation characteristics were sorted in descending order according to the regression coefficient, and the top three vegetation characteristics were selected as the vegetation characteristics with the strongest correlation with vegetation health.

[0075] In this invention, the three features with the strongest correlation with vegetation health are screened by linear regression analysis, which can effectively improve the accuracy and efficiency of the model, reduce the computational complexity, and enhance the interpretability and generalization ability of the model. It can provide more accurate health assessment and decision support for garden managers.

[0076] The normalization module 4 is used to normalize the characteristic values ​​of the three vegetation characteristics to between 0 and 255.

[0077] In a possible implementation, the normalization module 4 is specifically used to: use linear normalization to normalize the characteristic values ​​of the three vegetation characteristics to between 0 and 255:

[0078]

[0079] Among them, X′ represents the eigenvalue after normalization, X represents the eigenvalue before normalization, and Xmax Indicates the maximum value, X min Indicates the minimum value.

[0080] In the present invention, in image processing, the value of the RGB channel is usually represented by an integer between 0 and 255. By linearly normalizing the vegetation feature value to this range, the feature value can be mapped to the pixel value of the image, so that each feature directly corresponds to the color channel of the RGB image, which facilitates the conversion of data into a visual color vegetation feature map.

[0081] The fusion module 5 is used to fuse the first vegetation feature as the R channel, the second vegetation feature as the G channel, and the third vegetation feature as the B channel to form a color vegetation feature map.

[0082] In the present invention, the health status of vegetation often relies on the comprehensive evaluation of multiple features. By mapping multiple features to RGB channels respectively, the important information of each feature can be fused together, and information of multiple dimensions can be presented simultaneously in one image, which is convenient for comprehensive analysis. Furthermore, RGB images are a commonly used data format in convolutional neural networks. By fusing multiple features into RGB images, they can be directly used as input of CNN for automated health status analysis without the need for additional preprocessing steps.

[0083] The monitoring module 6 is used to monitor the health status of vegetation in the landscape garden by using the color vegetation feature map as input through a convolutional neural network.

[0084] Among them, Convolutional Neural Network (CNN) is a deep learning model used to process structured data such as images. It is particularly good at extracting features from images for classification, detection and recognition tasks. CNN is the main technology in the field of computer vision and can automatically learn complex features in images.

[0085] Furthermore, the convolutional neural network can automatically extract multi-level features from the color vegetation feature map, including edge, color, texture and other information, and gradually understand the complex structure in the image through different convolutional layers. Compared with traditional manual feature extraction methods, CNN can dig out more details and improve the ability to identify vegetation health problems.

[0086] In a possible implementation, the monitoring module 6 is specifically used for:

[0087] The input layer, the first convolution unit, the second convolution unit and the third convolution unit are connected in sequence, and the outputs of each convolution unit are cascaded to construct a convolutional neural network with a cascade structure.

[0088] Input the color vegetation feature map to the input layer.

[0089] A first feature map is extracted by the first convolution unit, a second feature map is extracted by the second convolution unit, and a third feature map is extracted by the third convolution unit.

[0090] In the present invention, by extracting features layer by layer through the first, second and third convolution units, the image can be parsed step by step from low-level features to high-level features. The low-level convolution layer extracts simple features (such as edges and lines), the middle level extracts textures and shapes, and the high level captures more abstract and complex features. This layered processing method can analyze the health status of vegetation from multiple angles and enhance the recognition ability of the model.

[0091] The first monitoring result is obtained by monitoring the health status of vegetation through the third feature map alone, the second monitoring result is obtained by cascading and fusing the second feature map and the third feature map, and the third monitoring result is obtained by monitoring the health status of vegetation through cascading and fusing the first feature map, the second feature map and the third feature map.

[0092] Optionally, the first monitoring result is specifically:

[0093] P1=Softmax[W1F3+b1]

[0094] P1={p 11 ,L p 1i ,L,p 1n}

[0095] Among them, P1 represents the first monitoring result, Softmax represents the Softmax activation function, W1 represents the first weight matrix, F3 represents the third feature map, b1 represents the first bias term, p 1i represents the probability value of belonging to the i-th health state in the first monitoring result, and n represents the total number of fault categories.

[0096] Optionally, the second monitoring result is specifically:

[0097]

[0098] P2={p 21 ,L p 2i ,L,p 2n}

[0099] Wherein, P2 represents the second monitoring result, W2 represents the second weight matrix, F2 represents the second feature map, represents element-by-element addition, represents the third feature map after upsampling. Upsampling is used to make the size of the feature map consistent. b2 represents the second bias term. p 2i Represents the probability value of belonging to the i-th health state in the second monitoring result.

[0100] Optionally, the third monitoring result is specifically:

[0101]

[0102] P3={p 31 ,L p 3i ,L,p 3n}

[0103] Wherein, P3 represents the third monitoring result, W3 represents the third weight matrix, F1 represents the first feature map, represents the second feature map after upsampling, b3 represents the third bias term, p 3i Represents the probability value of belonging to the i-th health state in the third monitoring result.

[0104] In the present invention, by cascading and fusing feature maps at different levels, the system can integrate information from different convolutional layers to achieve multi-level and multi-dimensional comprehensive analysis. The deeper the convolutional network, the more complex the extracted features. The combination of low-level local features and high-level global features can form a stronger expression ability, thereby improving the model's recognition effect on complex health problems. Furthermore, through the analysis of feature maps at different levels, the model can gradually reveal the health status of vegetation from basic to advanced. This layered monitoring method provides more information, making it easier for analysts to understand what features the system is based on to make judgments, thereby improving the interpretability of the model. For example, the first monitoring result may rely on simple shapes and edges, while the second and third monitoring results may combine more complex textures and patterns.

[0105] The first monitoring result, the second monitoring result and the third monitoring result are integrated to determine the final monitoring result.

[0106] Optionally, the final monitoring result is specifically:

[0107] p i =ω1p 1i +ω2p 2i +ω3p 3i

[0108] Among them, p i represents the probability value of belonging to the i-th health state, ω1 represents the weight coefficient of the first monitoring result, ω2 represents the weight coefficient of the second monitoring result, and ω3 represents the weight coefficient of the third monitoring result.

[0109] Among them, those skilled in the art can set the size according to actual conditions, and the present invention does not limit it.

[0110] The health state with the largest probability value is taken as the final monitoring result of this health state monitoring.

[0111] In the present invention, the features extracted by each layer of convolutional units reflect different health dimensions of vegetation (such as edges, textures, shapes, etc.), and the fusion of monitoring results at each layer can more comprehensively reflect the health status of vegetation. Low-level features often capture simple features, while high-level features can understand more complex global patterns. Combining this information can ensure that the monitoring results are more comprehensive and accurate.

[0112] Optionally, the health status includes: healthy, mildly stressed, moderately stressed, severely stressed, and dying.

[0113] Among them, in a healthy state, the vegetation grows normally without obvious signs of stress or damage. The leaves are rich green and photosynthesize efficiently. In a state of mild stress, the vegetation shows some mild signs of stress, such as lack of water, insufficient nutrients or temperature stress, but is in good overall health. In a state of moderate stress, the vegetation is obviously affected by stress, and the leaves may turn yellow or show some wilting. This may be caused by drought, disease or other environmental factors. In a state of severe stress, the vegetation is in a state of severe stress, with large areas of leaves turning yellow and falling off, and may be in a dying state. Common causes include severe pests and diseases, extreme climatic conditions or severe water shortage. In a dying state, the vegetation is dead or almost dead, completely losing signs of life activity, most of the leaves have fallen off, and the branches are dry.

[0114] Furthermore, alarm information can be issued when mild stress, moderate stress, severe stress and near-death occur, reminding garden management personnel to take measures immediately.

[0115] Furthermore, the focus of management measures varies according to the different health states of vegetation. For healthy and slightly stressed vegetation, the main focus is to maintain normal management and timely adjustments; for moderately and severely stressed vegetation, it is necessary to strengthen the management of water and nutrients, pay attention to pest control and pruning; for dying vegetation, it is usually necessary to remove and replant, and repair the planting environment. Reasonable management measures can effectively prevent the problem from getting worse and ensure that the vegetation in the landscape garden remains healthy and continues to grow.

[0116] In the present invention, the use of convolutional neural networks can realize automated vegetation health status detection. The system uses color vegetation feature maps as input, automatically extracts key features in the image, and completes health monitoring without human intervention. This greatly reduces labor costs and is particularly suitable for monitoring large-scale landscape gardens.

[0117] Furthermore, the improved Adam optimizer can be used to train the convolutional neural network.

[0118] The convolutional neural network is trained using the mean square error loss function and the improved Adam optimizer to adaptively adjust the learning rate and acceleration parameters:

[0119]

[0120] Among them, θ t represents the convolutional neural network parameters at the tth iteration, θ t-1 represents the convolutional neural network parameters at the t-1th iteration, α represents the basic learning rate, represents the second-order moment estimate after the correction of the deviation at the t-th iteration, ε represents the hyperparameter for numerical stability, represents the first-order moment estimate after the correction of the deviation at the tth iteration, v t represents the second-order moment estimate at the tth iteration, β2 represents the second-order moment estimate attenuation coefficient, and v t-1 represents the second-order moment estimate at the t-1th iteration, g t represents the gradient at the tth iteration, represents the gradient operation, f represents the loss function, Indicates that the loss function performs gradient calculation on the convolutional neural network parameters at the tth iteration, m t represents the first-order moment estimate at the t-th iteration, β1 represents the first-order moment estimate attenuation coefficient, m t-1 Represents the first-order moment estimate at the t-1th iteration.

[0121] It should be noted that the improved Adam optimizer adaptively adjusts the learning rate of each parameter based on the historical gradient information of each parameter. For parameters with drastic gradient changes, Adam will reduce the learning rate, and for parameters with gentle gradient changes, it will increase the learning rate. This adaptability enables the model to converge faster while reducing training problems caused by improper learning rate selection.

[0122] In the present invention, by using the improved Adam optimizer, the model training process can adaptively adjust the learning rate and acceleration parameters, effectively handle sparse gradients, improve numerical stability, accelerate convergence, and reduce sensitivity to hyperparameters. This optimization method is particularly suitable for complex convolutional neural networks, which can improve the training efficiency, stability and generalization ability of the model, and ultimately achieve better prediction performance.

[0123] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0124] In the present invention, through the landscape gardening environment monitoring device, a convolutional neural network is used to automatically monitor the health status of vegetation in all areas of the landscape garden, which can improve the monitoring efficiency of landscape gardening, provide more accurate and real-time health assessments, reduce labor costs and subjective errors, and realize intelligent and scientific garden management.

[0125] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A landscape gardening environment monitoring device, characterized in that: include: An acquisition module, used for acquiring image data of landscape architecture; An extraction module, used for extracting a plurality of vegetation features from the image data; A screening module is used to screen multiple vegetation features and determine the three vegetation features that are most closely related to vegetation health; The normalization module is used to normalize the characteristic values ​​of the three vegetation features to between 0 and 255; A fusion module, used to fuse the first vegetation feature as an R channel, the second vegetation feature as a G channel, and the third vegetation feature as a B channel to form a color vegetation feature map; The monitoring module is used to monitor the health status of vegetation in the landscape garden by using the color vegetation feature map as input through a convolutional neural network.

2. The landscape gardening environment monitoring device according to claim 1, characterized in that: The extraction module is specifically used for: Extract the green sum index SGI of each region from the conventional image; The enhanced vegetation index EVI, moisture stress index MSI, structure-insensitive pigment index SIPI, atmospheric resistance vegetation index ARVI, normalized difference vegetation index NDVI, green normalized difference vegetation index GNDVI and blue normalized difference vegetation index BNDVI of each area are extracted from infrared remote sensing images.

3. The landscape gardening environment monitoring device according to claim 1, characterized in that: The screening module is specifically used for: Through principal component analysis, multiple vegetation features were screened and the three vegetation features with the strongest correlation with vegetation health were identified.

4. The landscape gardening environment monitoring device according to claim 1, characterized in that: The screening module is specifically used for: Through linear regression analysis, multiple vegetation characteristics were screened and the three vegetation characteristics with the strongest correlation with vegetation health were determined.

5. The landscape gardening environment monitoring device according to claim 4, characterized in that: The screening module is specifically used for: Linear fitting is performed between multiple vegetation characteristics and vegetation health to determine the regression coefficient of each vegetation characteristic; The vegetation characteristics were sorted in descending order according to the regression coefficient, and the top three vegetation characteristics were selected as the vegetation characteristics with the strongest correlation with vegetation health.

6. The landscape gardening environment monitoring device according to claim 1, characterized in that: The normalization module is specifically used for: Linear normalization is used to normalize the characteristic values ​​of the three vegetation features to between 0 and 255.

7. The landscape gardening environment monitoring device according to claim 1, characterized in that: The monitoring module is specifically used for: The input layer, the first convolution unit, the second convolution unit, and the third convolution unit are sequentially connected, and the outputs of each convolution unit are cascaded to construct a convolutional neural network with a cascade structure; Inputting the color vegetation feature map into the input layer; Extracting a first feature map through the first convolution unit, extracting a second feature map through the second convolution unit, and extracting a third feature map through the third convolution unit; The third feature map is used to monitor the health status of vegetation alone to obtain a first monitoring result, the second feature map and the third feature map are cascaded and fused to monitor the health status of vegetation to obtain a second monitoring result, and the first feature map, the second feature map and the third feature map are cascaded and fused to monitor the health status of vegetation to obtain a third monitoring result; The first monitoring result, the second monitoring result and the third monitoring result are integrated to determine a final monitoring result.

8. The landscape gardening environment monitoring device according to claim 7, characterized in that: The health state with the largest probability value is taken as the final monitoring result of this health state monitoring.

9. The landscape gardening environment monitoring device according to claim 8, characterized in that: The health status includes: healthy, mild stress, moderate stress, severe stress and near death.

Images