Update app.py

app.py

@@ -1,112 +1,159 @@
-import ee
-import json
-import pandas as pd
 import streamlit as st
-import folium
-from branca.colormap import linear
+import pandas as pd
 import plotly.express as px
-import requests
+import folium
+from streamlit_folium import folium_static
+import ee
+import datetime
+import os
 
-# Authentication with Google Earth Engine
-credentials = ee.ServiceAccountCredentials('ee-esmaeilkiani13877-cfdea6eaf411.json', 'dehkhodamap-e9f0da4ce9f6514021@ee-esmaeilkiani13877.iam.gserviceaccount.com')
+# Set up Google Earth Engine
+service_account = 'dehkhodamap-e9f0da4ce9f6514021@ee-esmaeilkiani13877.iam.gserviceaccount.com'
+credentials = ee.ServiceAccountCredentials(service_account, 'ee-esmaeilkiani13877-cfdea6eaf411.json')
 ee.Initialize(credentials)
 
-# Define the Area of Interest (AOI)
-aoi = ee.Geometry.Rectangle([-118.806, 33.914, -118.770, 33.941])  # Example coordinates, adjust as necessary
-
-# Functions to calculate indices
-def calculate_ndvi(image):
-    return image.normalizedDifference(['B8', 'B4'])
-
-def calculate_ndwi(image):
-    return image.normalizedDifference(['B3', 'B8'])
-
-def calculate_ndre(image):
-    return image.normalizedDifference(['B8', 'B5'])
-
-def calculate_lai(image):
-    return image.expression('2.5 * (B8 - B4) / (B8 + B4 + 0.5)', {
-        'B8': image.select('B8'),
-        'B4': image.select('B4')
-    })
-
-def calculate_chl(image):
-    return image.expression('10 * (B8 - B4) / (B8 + B4)', {
-        'B8': image.select('B8'),
-        'B4': image.select('B4')
-    })
-
-# Use Sentinel-2 data
-sentinel2 = ee.ImageCollection('COPERNICUS/S2').filterBounds(aoi).filterDate('2023-01-01', '2023-12-31')
-ndvi = sentinel2.map(calculate_ndvi)
-ndwi = sentinel2.map(calculate_ndwi)
-ndre = sentinel2.map(calculate_ndre)
-lai = sentinel2.map(calculate_lai)
-chl = sentinel2.map(calculate_chl)
-
-# Load CSV files
-farm_coords = pd.read_csv('farm_coordinates.csv')
-daily_farm_data = pd.read_csv('پایگاه داده (1).csv', encoding='utf-8')
-
-# Weather API Key
-API_KEY = 'ed47316a45379e2221a75f813229fb46'
-
-def get_weather_data(latitude, longitude):
-    url = f"http://api.openweathermap.org/data/2.5/weather?lat={latitude}&lon={longitude}&appid={API_KEY}&units=metric"
-    response = requests.get(url)
-    return response.json()
-
-def create_map(ndvi_image):
-    map = folium.Map()
-    folium.TileLayer('OpenStreetMap').add_to(map)
-    map_bounds = aoi.getInfo()['coordinates'][0]
-    map.add_child(folium.raster_layers.ImageOverlay(
-        image=ndvi_image.getMapId()['tile_fetcher'].url_format,
-        bounds=map_bounds,
-        opacity=0.7
-    ))
-    colormap = linear.YlOrRd_09.scale(0, 1)
-    colormap.caption = 'NDVI'
-    colormap.add_to(map)
-    return map
-
-def generate_weekly_report(farm_name):
-    # Pseudo code for weekly comparison, you'd need to implement actual data retrieval
-    last_week_lai = lai.filterDate('last_week_start', 'last_week_end').mean().getInfo()
-    this_week_lai = lai.filterDate('this_week_start', 'this_week_end').mean().getInfo()
-    comparison = "LAI increased by" if this_week_lai > last_week_lai else "LAI decreased by"
-    st.write(f"Weekly Report for {farm_name}: {comparison} {abs(this_week_lai - last_week_lai)}")
-
-# Streamlit App
-st.title('Sugarcane Farm Monitoring')
-
-# Dropdown for farm selection
-farm = st.selectbox('Select Farm', farm_coords['name'])
-
-# Display NDVI Map
-st.write('NDVI Map')
-ndvi_map = create_map(ndvi.mean())
-st.components.v1.html(ndvi_map._repr_html_(), width=800, height=600)
-
-# Display Weather Data
-st.write('Weather Data')
-if farm in farm_coords['name'].values:
-    farm_data = farm_coords[farm_coords['name'] == farm].iloc[0]
-    weather_info = get_weather_data(farm_data['latitude'], farm_data['longitude'])
-    st.write(f"Temperature: {weather_info['main']['temp']}°C")
-    st.write(f"Humidity: {weather_info['main']['humidity']}%")
-    st.write(f"Wind Speed: {weather_info['wind']['speed']} m/s")
-else:
-    st.write("Farm not found in the dataset.")
-
-# Weekly Report
-st.write('Weekly Report')
-generate_weekly_report(farm)
-
-# Time Series Plot for NDVI
-st.write('NDVI Time Series')
-fig = px.line(ndvi.getRegion(aoi, 30).getInfo(), x='system:time_start', y='nd')
-st.plotly_chart(fig)
-
-# Note: This script assumes you have the necessary permissions, setup, and libraries installed. 
-# Also, replace placeholder data with real-time data where applicable.
+# Load farm data
+@st.cache_data
+def load_farm_data():
+    return pd.read_csv('farm_coordinates.csv')
+
+farms = load_farm_data()
+
+# Streamlit app
+st.title('Sugarcane Farm Dashboard')
+
+# Sidebar
+st.sidebar.title('Navigation')
+page = st.sidebar.radio('Go to', ['Map', 'Comparative Charts', 'Weekly Reports'])
+
+if page == 'Map':
+    st.header('Sugarcane Farm Map')
+    
+    # Create map
+    m = folium.Map(location=[farms['latitude'].mean(), farms['longitude'].mean()], zoom_start=10)
+    
+    # Add markers for each farm
+    for idx, row in farms.iterrows():
+        folium.Marker(
+            location=[row['latitude'], row['longitude']],
+            popup=f"Farm: {row['name']}, Age: {row['age']}, Variety: {row['variety']}",
+            icon=folium.Icon(color='green', icon='leaf')
+        ).add_to(m)
+    
+    # Display map
+    folium_static(m)
+    
+    # Farm search
+    search_farm = st.text_input('Search for a farm:')
+    if search_farm:
+        farm = farms[farms['name'].str.contains(search_farm, case=False)]
+        if not farm.empty:
+            st.write(f"Farm found: {farm['name'].values[0]}")
+            st.map(farm)
+        else:
+            st.write("Farm not found.")
+
+elif page == 'Comparative Charts':
+    st.header('Comparative Charts')
+    
+    # Function to get NDVI data
+    def get_ndvi_data(geometry, start_date, end_date):
+        collection = ee.ImageCollection('COPERNICUS/S2_SR') \
+            .filterBounds(geometry) \
+            .filterDate(start_date, end_date) \
+            .filter(ee.Filter.lt('CLOUDY_PIXEL_PERCENTAGE', 20))
+        
+        def addNDVI(image):
+            ndvi = image.normalizedDifference(['B8', 'B4']).rename('NDVI')
+            return image.addBands(ndvi)
+        
+        ndvi_collection = collection.map(addNDVI)
+        ndvi_values = ndvi_collection.select('NDVI').mean().reduceRegion(
+            reducer=ee.Reducer.mean(),
+            geometry=geometry,
+            scale=10
+        ).get('NDVI')
+        
+        return ndvi_values.getInfo()
+    
+    # Function to get LAI data
+    def get_lai_data(geometry, start_date, end_date):
+        collection = ee.ImageCollection('MODIS/006/MOD15A2H') \
+            .filterBounds(geometry) \
+            .filterDate(start_date, end_date)
+        
+        lai_values = collection.select('Lai_500m').mean().reduceRegion(
+            reducer=ee.Reducer.mean(),
+            geometry=geometry,
+            scale=500
+        ).get('Lai_500m')
+        
+        return lai_values.getInfo()
+    
+    # Date range selection
+    start_date = st.date_input('Start date', datetime.date(2023, 1, 1))
+    end_date = st.date_input('End date', datetime.date.today())
+    
+    if start_date < end_date:
+        # Get NDVI and LAI data for each farm
+        ndvi_data = []
+        lai_data = []
+        for idx, row in farms.iterrows():
+            geometry = ee.Geometry.Point([row['longitude'], row['latitude']])
+            ndvi = get_ndvi_data(geometry, start_date.isoformat(), end_date.isoformat())
+            lai = get_lai_data(geometry, start_date.isoformat(), end_date.isoformat())
+            ndvi_data.append({'Farm': row['name'], 'NDVI': ndvi})
+            lai_data.append({'Farm': row['name'], 'LAI': lai})
+        
+        ndvi_df = pd.DataFrame(ndvi_data)
+        lai_df = pd.DataFrame(lai_data)
+        
+        # NDVI chart
+        fig_ndvi = px.bar(ndvi_df, x='Farm', y='NDVI', title='NDVI Comparison')
+        st.plotly_chart(fig_ndvi)
+        
+        # LAI chart
+        fig_lai = px.bar(lai_df, x='Farm', y='LAI', title='LAI Comparison')
+        st.plotly_chart(fig_lai)
+    else:
+        st.error('Error: End date must fall after start date.')
+
+elif page == 'Weekly Reports':
+    st.header('Weekly Reports')
+    
+    # Farm selection
+    selected_farm = st.selectbox('Select a farm:', farms['name'])
+    
+    # Week selection
+    week = st.slider('Select week:', 1, 52, 1)
+    
+    # Simulate weekly data (replace this with actual data processing)
+    def get_weekly_data(farm, week):
+        return {
+            'NDVI': 0.5 + 0.1 * week,  # Simulated increasing NDVI
+            'LAI': 2 + 0.2 * week,  # Simulated increasing LAI
+        }
+    
+    data = get_weekly_data(selected_farm, week)
+    
+    # Display weekly report
+    st.subheader(f'Weekly Report for {selected_farm} - Week {week}')
+    
+    col1, col2 = st.columns(2)
+    with col1:
+        st.metric(label="NDVI 🌱", value=f"{data['NDVI']:.2f}")
+    with col2:
+        st.metric(label="LAI 🌡", value=f"{data['LAI']:.2f}")
+    
+    # Trend chart
+    weeks = list(range(1, week + 1))
+    trend_data = [get_weekly_data(selected_farm, w) for w in weeks]
+    trend_df = pd.DataFrame(trend_data)
+    trend_df['Week'] = weeks
+    
+    fig = px.line(trend_df, x='Week', y=['NDVI', 'LAI'], title=f'Trend for {selected_farm}')
+    st.plotly_chart(fig)
+
+# Footer
+st.sidebar.markdown('---')
+st.sidebar.text('Dashboard created with Streamlit')