常规更新

src/gasflux/pre_processing.py

@@ -1,11 +1,9 @@
 """Functions that organise the data into standard columns in pandas dataframes. Conversion functions (e.g. WGS84 to UTM)
 are here but transformations take place in processing.py"""
 
-import geopandas as gpd
 import numpy as np
 import pandas as pd
-
-from . import plotting
+from pyproj import Transformer
 from .processing import circ_median
 
 
@@ -37,19 +35,52 @@ def timestamp_from_four_columns(df):
 
 # add UTM from latitudes and longitudes
 def add_utm(df: pd.DataFrame) -> pd.DataFrame:
-    gdf = gpd.GeoDataFrame(  # type: ignore
-        df,
-        geometry=gpd.points_from_xy(df["longitude"], df["latitude"], crs="EPSG:4326"),
-    )
-    utm = gdf.estimate_utm_crs()
-    gdf = gdf.to_crs(utm)
-    if not isinstance(gdf, gpd.GeoDataFrame):
-        raise TypeError("Failed to reproject to a GeoDataFrame")
-    gdf["utm_easting"] = gdf.geometry.x
-    gdf["utm_northing"] = gdf.geometry.y
-    output_df = pd.DataFrame(gdf.drop(columns="geometry"))
+    """
+    Convert WGS84 coordinates to UTM using pyproj.
 
-    return output_df
+    This function replaces the geopandas implementation with a lighter pyproj-based solution.
+    """
+    # Make a copy to avoid modifying the original DataFrame
+    df = df.copy()
+
+    # Get the average longitude to determine the UTM zone
+    # For simplicity, we'll use the first valid longitude to determine the zone
+    # In production, you might want to use the centroid or handle multiple zones
+    valid_lons = df["longitude"].dropna()
+    if len(valid_lons) == 0:
+        raise ValueError("No valid longitude values found")
+
+    # Calculate UTM zone from longitude
+    # UTM zones are 6 degrees wide, starting from -180
+    lon = valid_lons.iloc[0]  # Use first valid longitude
+    zone_number = int((lon + 180) / 6) + 1
+
+    # Determine if it's northern or southern hemisphere
+    # Use first valid latitude
+    valid_lats = df["latitude"].dropna()
+    if len(valid_lats) == 0:
+        raise ValueError("No valid latitude values found")
+
+    lat = valid_lats.iloc[0]
+    hemisphere = 'north' if lat >= 0 else 'south'
+
+    # Create UTM CRS string
+    utm_crs = f"EPSG:326{zone_number:02d}" if hemisphere == 'north' else f"EPSG:327{zone_number:02d}"
+
+    # Create transformer from WGS84 to UTM
+    transformer = Transformer.from_crs("EPSG:4326", utm_crs, always_xy=True)
+
+    # Transform coordinates
+    utm_easting, utm_northing = transformer.transform(
+        df["longitude"].values,
+        df["latitude"].values
+    )
+
+    # Add UTM coordinates to DataFrame
+    df["utm_easting"] = utm_easting
+    df["utm_northing"] = utm_northing
+
+    return df
 
 
 # add columns for drone course azimuth and elevation
@@ -97,7 +128,9 @@ def remove_outliers(df: pd.DataFrame, column: str, name: str):
     iqr = q3 - q1
     fence_low = q1 - 3 * iqr
     fence_high = q3 + 3 * iqr
-    fig = plotting.outliers(df[column], fence_high, fence_low)
+
+    fig = None  # plotting disabled
+
     outliers = df.loc[(df[column] < fence_low) | (df[column] > fence_high)]
     if len(outliers) > 0:
         print(f"{len(outliers)} outliers removed from {name} {column} data")