Tropical cyclone-based mangrove health assessment system

Abstract

The present application relates to the technical field of data analysis, in particular to a mangrove health assessment system based on tropical cyclones. The health risk index of the mangrove is obtained by combining the structural risk coefficient, the geological risk coefficient and the ecological risk coefficient, thereby realizing accurate health assessment. In addition, meteorological data and cyclone state data are monitored to predict the cyclone prediction path. Then, the cyclone prediction path and the health risk index of the mangrove are combined to realize the risk response control of damaged and undamaged mangroves, so as to form accurate judgment and supervision of key areas, thereby reducing the excessive damage of mangroves due to untimely control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, in particular to a mangrove health evaluation system based on tropical cyclones. BACKGROUND

[0002] In the growth process of mangroves, if they are in the moving path of cyclones, they will be affected by cyclones. Therefore, after the mangrove area is affected, in order to analyze the health status of the mangrove, the current machine learning model and convolutional neural network model can be used to collect image data of trees in the mangrove area, and the image data can be labeled and trained to analyze the health degree of the mangrove. However, the above-mentioned analysis of the health degree may lead to inaccurate evaluation of the health degree of the mangrove, because the health status of the mangrove is not only reflected in the image data on its surface, but also affected by the environmental state.

[0003] Therefore, in order to improve the evaluation and analysis of the health degree of the mangrove after the cyclone, so as to effectively control the risk of the damaged and undamaged (in the moving path of the cyclone and the position not reached by the cyclone) mangroves, the present application provides a mangrove health evaluation system based on tropical cyclones. SUMMARY

[0004] In view of the above-mentioned shortcomings of the prior art, the present application provides a mangrove health evaluation system based on tropical cyclones, which can effectively solve the problem of inaccurate evaluation of the health degree of the mangrove in the prior art, and affect the early warning and control of the damaged and undamaged mangroves.

[0005] To achieve the above purpose, the present application is realized by the following technical scheme:

[0006] The present application provides a mangrove health evaluation system based on tropical cyclones, which comprises a health state analysis module for obtaining the state of the mangrove area affected by the cyclone and dividing the mangrove area into a same period mangrove area, obtaining the health risk index of the same period mangrove area , wherein:

[0007] Health risk index According to the structural risk coefficient , the geological risk coefficient and the ecological risk coefficient are combined to obtain;

[0008] According to the health risk index , the health degree of the same period mangrove area is determined;

[0009] Further comprising:

[0010] The correlation risk monitoring module is configured to acquire real-time meteorological data and cyclone state data, output a cyclone prediction path using a WRF model, mark a same-period mangrove area under the cyclone prediction path as a risk mangrove area, and predict a health risk index of the risk mangrove area according to a health degree of the same-period mangrove area Thus, it is determined whether to generate an emergency plan and execute.

[0011] The structural risk coefficient

[0012] The structural risk coefficient of the same-period mangrove area is acquired The structural risk coefficient , a slope difference value , a damaged area , and a damage rate are normalized and calculated according to the following relationship:

[0013]

[0014] In the formula: is a height difference value of the mangrove, is a slope difference value of the mangrove, is a damage rate, is a damaged area of the mangrove, , , and are corresponding weight coefficients, respectively, is a calculation item number, is a constant correction coefficient.

[0015] The geological risk coefficient

[0016] The geological risk coefficient of the same-period mangrove area is acquired The geological risk coefficient , a soil moisture difference value , an organic matter difference value , a soil difference value , and a soil structure value are normalized and calculated according to the following relationship:

[0017]

[0018] In the formula: is a soil erosion value, is a soil moisture difference value, is an organic matter difference value, is a soil difference value, is a soil structure value. , , , as well as These are the corresponding weight coefficients. To calculate the number of items, This is a constant correction factor.

[0019] The soil structure value The method for determining it is as follows:

[0020] The porosity, permeability, stability, and bulk density of soil in each mangrove area during the same period were pre-acquired and normalized to obtain the soil structure value of the soil in the current mangrove area during the same period. The soil structure value of all mangrove areas during the same period was obtained. The calculation formula for the soil structure value is as follows:

[0021]

[0022] In the formula, For soil porosity, For soil permeability, For soil stability, This refers to the volume density of the soil. , , as well as These are all corresponding weighting coefficients.

[0023] The ecological risk coefficient

[0024] Obtain the ecological risk coefficient of the mangrove area during the same period Ecological risk coefficient Water level difference value Salinity difference value and ecological pollution values Perform normalization and calculate according to the following formula:

[0025]

[0026] In the formula: This represents the tidal difference value. This represents the difference in water level. This represents the salinity difference value. This represents the ecological pollution value. , , and These are the corresponding weight coefficients. To calculate the number of items, This is a constant correction factor.

[0027] The health risk index

[0028]

[0029] In the formula: , and These are the weighting coefficients, This is the number of items to be counted.

[0030] The formula for judging the health status of the mangrove area during the same period is:

[0031]

[0032] like When the result is A, the current health status of the mangrove area is considered to be good, and no control measures are required.

[0033] Conversely, if If the result is not A, the current health status of the mangrove area is deemed poor, and control measures are required.

[0034] The meteorological data includes wind speed, wind direction, air pressure, temperature, and humidity;

[0035] Cyclone status data includes the cyclone's location, intensity, and size.

[0036] When the cyclone prediction path is determined:

[0037] By obtaining the coordinates of the high-risk mangrove areas and the current coordinates of the cyclone, the travel time of the cyclone to different high-risk mangrove areas can be calculated. ;

[0038] travel time With the current real-time Combined with the predicted time when the cyclone is in the risk mangrove zone ;

[0039] Obtain environmental and mangrove growth data for high-risk mangrove areas. Identify high-risk mangrove areas within contemporaneous mangrove areas affected by cyclones that match the environmental, mangrove growth, and cyclone risk data. Predict the health risk index of these high-risk mangrove areas. ,in:

[0040] If a health risk index exists in the mangrove area If the mangrove area is classified as B, C, or D, then that risky mangrove area is marked as the target mangrove area, based on the health risk index. The corresponding health status generates the corresponding emergency plan.

[0041] When the target mangrove region is determined:

[0042] Define the predicted arrival time of cyclones in the target mangrove area. Obtain the evacuation time of personnel after the emergency plan is implemented. This generates the maximum completion time for emergency response plans corresponding to different target mangrove areas. To complete in the longest time Output the execution time of the emergency response plan to the control terminal.

[0043] The technical solution provided by this invention has the following advantages compared with the known prior art:

[0044] By combining structural risk coefficients, geological risk coefficients, and ecological risk coefficients to obtain a health risk index for mangroves, a precise health assessment can be achieved. Based on this, meteorological data and cyclone status data can be monitored to predict the movement path of cyclones. Furthermore, by combining the predicted cyclone path with the mangrove health risk index, risk response management can be implemented for both damaged and undamaged mangroves. This allows for precise identification and supervision of key areas, thereby reducing the risk of excessive damage to mangroves due to untimely management. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0046] Figure 1 This is a schematic diagram of the overall process of the present invention. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0048] The present invention will be further described below with reference to embodiments.

[0049] Example 1 (see Figure 1 A mangrove health assessment system based on tropical cyclones includes at least:

[0050] The health status analysis module is used to identify mangrove areas currently affected by cyclones. The specific steps for identifying and acquiring mangrove areas affected by cyclones are as follows:

[0051] Step 1: Collect remote sensing image data of mangrove areas before and after the cyclone using satellite imagery;

[0052] Step 2: Preprocess the remote sensing impact data, such as performing standardization on the remote sensing image data. The calculation formula is as follows:

[0053]

[0054] In the formula, For the processed remote sensing image data, The original remote sensing image data, The mean, Standard deviation;

[0055] Step 2: Extract relevant features from remote sensing image data, including vegetation indices (NDVI, EVI), land cover classification (mangroves, mudflats), and spectral features, to reflect the growth status of mangrove areas;

[0056] Step 3: Select samples from both affected and unaffected mangrove areas, label these samples, and define the training and testing samples. The training samples are used to train the machine learning model, while the testing samples are used to evaluate the machine learning model and optimize its parameters. The machine learning model used can be a convolutional neural network, decision tree, etc., and the training and testing samples can be used to test and optimize the model.

[0057] If a convolutional neural network is used for training and recognition, the steps are as follows:

[0058] Perform convolution operations:

[0059]

[0060] In the formula: For the first Layer Convolutional output at each position, For the first Layer The convolution kernel weights at each position, For the next level , The activation value of each position. For the first The offset top of the layer;

[0061] Execute the activation function:

[0062]

[0063] In the formula: For the first The output feature map of the layer;

[0064] Perform max pooling operation:

[0065]

[0066] The first element in the output feature map after pooling The values ​​at each position, where A is the input feature map. This is the row index of the current pooling window in the input feature map. This is the column index of the current pooling window in the input feature map;

[0067] Execute the output of the fully connected layer:

[0068]

[0069] In the formula: For the first The activation value of the layer, For the first Layer weight matrix;

[0070] Output of the Softmax layer:

[0071]

[0072] In the formula: For the model in the first Linear output of the class, For the model in the first Linear output of the class, For the sample Category The predicted probability, Given the total number of categories, calculate the probability of each category.

[0073] Execute the cross-entropy loss function:

[0074]

[0075] In the formula: For the first The true label of the sample in the th... The value of the class, for The sample at the th Predicted probability of class The total number of samples is used to calculate the error between the model and the true label.

[0076] Parameter update formula to adjust model parameters:

[0077]

[0078] In the formula: These are the model parameters, including all weights and biases. For learning rate, For the loss function with respect to the parameters The gradient represents the rate of change of the loss function in the parameter space;

[0079] Step 4: Based on the machine learning model (i.e., convolutional neural network model) constructed above, the trained machine learning model (convolutional neural network model) is applied to the identification area to identify and output the affected mangrove area through machine learning.

[0080] Therefore, based on the growth cycle, a large number of mangrove areas were divided to identify mangrove areas with the same growth cycle (growth time) (mangrove areas with the same growth cycle or growth time were marked to facilitate further analysis of the mangrove situation). The health risk index of these mangrove areas was then calculated. Among these, the health risk index Based on structural risk coefficient Geological risk coefficient and ecological risk coefficient The specific calculation steps are as follows:

[0081] Step 1: Obtain the structural risk coefficient of the mangrove area during the same period. Structural risk coefficient Based on height difference value Slope difference value Damaged area and damage rate Perform normalization and calculate according to the following relationship:

[0082]

[0083] In the formula: To obtain the height difference value of mangroves, the difference between the average tree height of all mangroves in the current mangrove area and the historical average tree height is obtained in advance (the corresponding historical mangrove area (the historical mangrove area whose growth parameters meet the preset standard values) is obtained; the historical average diameter is obtained in the same way, including steps two and three, etc., which will not be repeated later). The average height difference (height difference value) of mangroves is obtained based on the number of the same mangrove area in the same period.

[0084] To obtain the slope difference value of mangroves, the slope difference between the average slope of all mangroves in the current mangrove area and the historical average slope is obtained in advance, and the slope difference value (slope difference value) of mangroves is obtained based on the number of mangrove areas in the same period.

[0085] To determine the damage rate, the total number of damaged mangroves in all mangrove areas during the same period (damage includes mangrove diameters, branches with broken, bent, scratched, tilted, and collapsed states) is obtained, and the ratio of this number to the total number of mangroves is used to obtain the damage rate.

[0086] To determine the damaged area of ​​mangroves, obtain the average damaged area of ​​all mangroves in the current mangrove area (including obtaining the sum of the fracture area, scratch area, and bark peeling area of ​​mangroves in each mangrove area in the current period, in order to obtain the average damaged area of ​​multiple mangrove areas in the current period). , , and These are the corresponding weight coefficients. The default value for the number of items to be calculated is 3. This is a constant correction factor.

[0087] Step 2: Obtain the geological risk coefficient of the mangrove area during the same period. Geological risk coefficient Based on soil erosion value Soil moisture difference value Organic matter difference value ,soil Difference value and soil structure values Perform normalization and calculate according to the following formula:

[0088]

[0089] In the formula: This represents the soil erosion value, based on the current volume of eroded soil in all mangrove areas during the same period (soil volume). Soil area The ratio of soil depth to total soil volume is used to obtain the result.

[0090] To obtain the soil moisture difference value, the difference between the current soil moisture and the historical soil moisture in the mangrove area is obtained in advance, and the average moisture difference (soil moisture difference value) of all mangrove areas in the same period is obtained. Here, the soil moisture is obtained according to the ratio of soil water content to soil volume.

[0091] To obtain the organic matter difference value, the difference between the current organic matter content of the mangrove soil and the historical organic matter content is obtained in advance, and the average content difference (organic matter difference value) of all mangrove areas in the same period is obtained.

[0092] For soil Difference values, obtained in advance from the soil samples of the mangrove area during the same period. The soil of value and history The pH difference between values ​​was used to obtain the average pH difference (soil) of all mangrove areas during the same period. (difference value);

[0093] To obtain the soil structure value, the porosity, permeability, stability, and bulk density of the soil in each mangrove area during the same period are pre-acquired and normalized to obtain the soil structure value of the current mangrove area during the same period. The average soil structure value of all mangrove areas during the same period is then obtained, which is the soil structure value. The formula for calculating the soil structure value is as follows:

[0094]

[0095] In the formula, For soil porosity, For soil permeability, Soil stability is determined by the product of the number of aggregates, their geometric mean diameter, and their average mass diameter. This refers to the volume density of the soil. , , , as well as These are the corresponding weight coefficients. The default value for the number of items to be calculated is 4. The constant correction factor is used. , , as well as These are all corresponding weighting coefficients.

[0096] Step 3: Obtain the ecological risk coefficient of the mangrove area during the same period. Ecological risk coefficient Based on tidal difference value Water level difference value Salinity difference value and ecological pollution values Perform normalization and calculate according to the following formula:

[0097]

[0098] In the formula: To obtain the tidal difference value, the tidal difference between the current tidal height and the historical tidal height in the mangrove area is obtained in advance, and the average tidal difference value (tidal difference value) of all mangrove areas in the same period is obtained.

[0099] The value for water level difference is determined by whether the average water level change in all mangrove areas during the same period falls within the historical water level change range. If it does, the value is 0.1; otherwise, it is 0.5.

[0100] To obtain the salinity difference value, the salinity difference between the current salinity and the historical salinity in the mangrove area is obtained in advance, and the average salinity difference (salinity difference value) of all mangrove areas in the same period is obtained.

[0101] To determine the ecological pollution value, the product between the heavy metal content and the organic pollutant concentration in each mangrove area during the same period is obtained in advance, thus clarifying the pollution value of each mangrove area during the same period. The average pollution value of all mangrove areas during the same period is then obtained, which is the ecological pollution value. , , and These are the corresponding weight coefficients. The default value for the number of items to be calculated is 4. This is a constant correction factor.

[0102] Step 4: Calculate the health risk index :

[0103]

[0104] In the formula: , and These are the weighting coefficients, The number of items to be calculated is set to 3 by default.

[0105] It should be noted that in this scheme, the weight coefficients can be determined by the analytic hierarchy process or the principal component analysis method.

[0106] Principal Component Analysis (PCA): Identifies the main influencing factors by dimensionality reduction and assigns weights based on their contribution.

[0107] Analytic Hierarchy Process (AHP): Constructs a judgment matrix and calculates weights by comparing the importance of each parameter. As mentioned above, both the analytic hierarchy process and principal component analysis are well-known techniques, and will not be elaborated further in this case.

[0108] Therefore, based on the above steps, the current health risk index of the mangrove area is obtained. Furthermore, a preset health risk level is obtained to determine the health status of the mangrove area after being affected by the cyclone during the same period. The formula for the determination is as follows:

[0109]

[0110] In the above, if it is A, it means that the current mangrove area is in good health and the impact of the cyclone on the mangrove area can be ignored.

[0111] Conversely, if it is not A, it indicates that the current mangrove area is in poor health, and the cyclone has a certain impact on the mangrove area, requiring control and management measures.

[0112] The associated risk monitoring module acquires real-time meteorological data, including wind speed, wind direction, air pressure, temperature, and humidity, and cyclone state data, including the cyclone's location, intensity, and size. This cyclone state data and meteorological data are input into the WRF model to output predicted cyclone paths at different future times. Furthermore, it determines whether mangrove areas exist along the predicted future cyclone paths. If so, these mangrove areas are marked as risky mangrove areas. From this, the coordinates of the risky mangrove areas and the current coordinates of the cyclone are obtained. Get the travel time of cyclones to mangrove areas with different risks. This will extend the travel time. With the current real-time By combining these methods, we can obtain the predicted time when the cyclone will be in the high-risk mangrove zone. The method involves acquiring environmental and mangrove growth data for high-risk mangrove areas. Based on this data (mangrove growth cycle time) and cyclone risk data (intensity and size of cyclones at different locations as output by the WRF model), high-risk mangrove areas matching (identical to) the environmental, mangrove growth, and cyclone risk data from mangrove areas previously affected by cyclones are identified. Following this, the health risk index of these high-risk mangrove areas can be directly predicted. ,in:

[0113] If a health risk index exists in the mangrove area If the situation is A, then the emergency plan will not be implemented for the risky mangrove area (generally speaking, the emergency plan is preset in the database based on the degree of damage to the risky mangrove area and the intensity and size of the cyclone).

[0114] If a health risk index exists in the mangrove area If the mangrove area is classified as B, C, or D, then that risky mangrove area is marked as the target mangrove area, based on the health risk index. The corresponding health level generates a corresponding emergency plan to reduce the damage to the target mangrove area, thereby improving the health level of the target mangrove area.

[0115] It is worth noting that the emergency plan includes:

[0116] Tree reinforcement: For important trees or those located in dangerous areas, reinforcement measures such as adding supports, tying, or pruning branches are taken to improve their ability to withstand wind.

[0117] Preventive pruning: Preventive pruning of trees that may be affected, removing dead, diseased, and overly long branches that may break due to strong winds, to reduce tree damage caused by strong winds.

[0118] Temporary shelter measures: Before the cyclone arrives, set up temporary shelter facilities, such as windbreaks or shade nets, for vulnerable tree areas to reduce the direct effect of wind.

[0119] Remove vulnerable trees: Vulnerable trees or those located in high-risk areas can be removed in advance to prevent further damage caused by tree collapse.

[0120] It should be noted that when identifying the target mangrove area, the predicted arrival time of the cyclone in the target mangrove area should be specified. Obtain the evacuation time of personnel after the emergency plan is implemented. (Time taken to evacuate the target mangrove area), in order to predict the time. Consider evacuation time (Predicted time) Reduce evacuation time (This is used to generate the maximum completion time of emergency plans for different target mangrove areas.) To complete in the longest time Output to the control terminal to indicate the execution time of the emergency plan corresponding to different target mangrove areas, so as to facilitate the safe execution of the emergency plan.

[0121] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A mangrove health assessment system based on tropical cyclones, comprising a health status analysis module for acquiring the current status of mangrove areas affected by cyclones and dividing the mangrove areas into concurrent mangrove zones, characterized in that, Obtain the health risk index corresponding to the mangrove area during the same period. ,in: Health Risk Index Based on structural risk coefficient Geological risk coefficient and ecological risk coefficient Combined to obtain; Based on the aforementioned health risk index Assess the health status of the mangrove area during the same period; Also includes: The associated risk monitoring module is used to acquire real-time meteorological and cyclone status data, output cyclone prediction paths using a WRF model, and designate mangrove areas within the predicted cyclone path as high-risk mangrove areas. Furthermore, it predicts the health risk index of high-risk mangrove areas based on the health status of other mangrove areas in the same period. This determines whether an emergency plan should be generated and implemented. The structural risk coefficient The method for determining it is as follows: Obtain the structural risk coefficient of the mangrove area during the same period Structural risk coefficient Based on height difference value Slope difference value Damaged area and damage rate Perform normalization and calculate according to the following relationship: ; In the formula: This represents the height difference value of mangroves. This represents the difference in slope of the mangroves. For damage rate, This represents the area of ​​damage to the mangroves. , , and These are the corresponding weight coefficients. The number of items used in calculating the structural risk coefficient. This is a constant correction factor; The geological risk coefficient The method for determining it is as follows: Obtain the geological risk coefficient of the mangrove area during the same period Geological risk coefficient Based on soil erosion value Soil moisture difference value Organic matter difference value ,soil Difference value and soil structure values Perform normalization and calculate according to the following formula: ; In the formula: This represents the soil erosion value. This represents the difference in soil moisture. This represents the difference in organic matter. For soil Difference value, This represents the soil structure value. , , , as well as These are the corresponding weight coefficients. This represents the number of items used in calculating the geological risk coefficient. This is a constant correction factor; The ecological risk coefficient The method for determining it is as follows: Obtain the ecological risk coefficient of the mangrove area during the same period Ecological risk coefficient Based on tidal difference value Water level difference value Salinity difference value and ecological pollution values Perform normalization and calculate according to the following formula: ; In the formula: This represents the tidal difference value. This represents the difference in water level. This represents the salinity difference value. This represents the ecological pollution value. , , and These are the corresponding weight coefficients. This represents the number of items used in calculating the ecological risk coefficient. This is a constant correction factor.

2. The mangrove health assessment system based on tropical cyclones according to claim 1, characterized in that, The soil structure value The method for determining it is as follows: The porosity, permeability, stability, and bulk density of soil in each mangrove area during the same period were pre-acquired and normalized to obtain the soil structure value of the soil in the current mangrove area during the same period. The soil structure value of all mangrove areas during the same period was obtained. The calculation formula for the soil structure value is as follows: ; In the formula, For soil porosity, For soil permeability, For soil stability, This refers to the volume density of the soil. , , as well as These are all corresponding weighting coefficients.

3. The mangrove health assessment system based on tropical cyclones according to claim 1, characterized in that, The health risk index The calculation formula is: ; In the formula: , and These are the weighting coefficients, This represents the number of items used in the calculation of the health risk index.

4. The mangrove health assessment system based on tropical cyclones according to claim 1, characterized in that, The formula for judging the health status of the mangrove area during the same period is: ; like When the result is A, the current health status of the mangrove area is considered to be good, and no control measures are required. Conversely, if If the result is not A, the current health status of the mangrove area is deemed poor, and control measures are required.

5. The mangrove health assessment system based on tropical cyclones according to claim 1, characterized in that, The meteorological data includes wind speed, wind direction, air pressure, temperature, and humidity; Cyclone status data includes the cyclone's location, intensity, and size.

6. The mangrove health assessment system based on tropical cyclones according to claim 5, characterized in that, When the cyclone prediction path is determined: By obtaining the coordinates of the high-risk mangrove areas and the current coordinates of the cyclone, the travel time of the cyclone to different high-risk mangrove areas can be calculated. ; travel time With the current real-time Combined with the predicted time when the cyclone is in the risk mangrove zone ; Obtain environmental and mangrove growth data for high-risk mangrove areas. Identify high-risk mangrove areas within contemporaneous mangrove areas affected by cyclones that match the environmental, mangrove growth, and cyclone risk data. Predict the health risk index of these high-risk mangrove areas. ,in: If a health risk index exists in the mangrove area If the mangrove area is classified as B, C, or D, then that risky mangrove area is marked as the target mangrove area, based on the health risk index. The corresponding health status generates the corresponding emergency plan.

7. The mangrove health assessment system based on tropical cyclones according to claim 6, characterized in that, When the target mangrove region is determined: Define the predicted arrival time of cyclones in the target mangrove area. Obtain the evacuation time of personnel after the emergency plan is implemented. This generates the maximum completion time for emergency response plans corresponding to different target mangrove areas. To complete in the longest time Output the execution time of the emergency response plan to the control terminal.

Images