Merge pull request #15 from 2320sharon/plot_geoaccuracy

Add plot_geoaccuracy.py Script

src/sdstools/plot_geoaccuracy.py

@@ -0,0 +1,132 @@
+"""
+Sharon Fitzpatrick Batiste
+1/16/2024
+This script plots the geo-reference accuracy for each image from any of the landsat satellites that have this metric available.
+It reads each satellite directory's metadatafolder and extracts the georeference accuracy for each image.
+Then plots each of these accuracies grouped by what month the image was captured and colored by the satellite it was captured by. 
+Currently works for Landsat 7,8,9.
+
+Expects the provided directory to be in format and each filename containing the metadata files to contain the date the image was captured 
+in the format YYYY-MM-DD-hh-mm-ss_satellite_sessionname.txt . This is the default format from coastsat and coastseg style downloads.
+
+
+sample_session_name
+|_L7
+  |_meta
+    |_2023-12-06-13-25-32_L7_sample_session_name.txt
+    |_2023-12-07-13-25-32_L7_sample_session_name.txt
+|_L8
+  |_meta
+    |_2023-12-06-13-25-32_L8_sample_session_name.txt
+    |_2023-12-07-13-25-32_L8_sample_session_name.txt
+|_L9
+  |_meta
+    |_2023-12-06-13-25-32_L9_sample_session_name.txt
+    |_2023-12-07-13-25-32_L9_sample_session_name.txt
+
+
+Expects the metadata text files to be in a format like the following example below:
+filename	2023-12-11-13-28-02_L7_ID_wra5_datetime01-10-25__03_58_44_ms.tif
+epsg	32617
+acc_georef	11.537
+image_quality	9
+im_width	335
+im_height	334
+
+"""
+import os
+import glob
+import re
+import pandas as pd
+import matplotlib.pyplot as plt
+import os
+import glob
+import re
+import pandas as pd
+import matplotlib.pyplot as plt
+
+def read_data_from_file(file_path):
+    """ Extract filename and acc_georef from a single file. """
+    data = {}
+    with open(file_path, "r") as file:
+        for line in file:
+            if line.startswith("filename"):
+                parts = line.split()
+                if len(parts) >= 2:
+                    data['filename'] = parts[1]
+            if line.startswith("acc_georef"):
+                parts = line.split()
+                if len(parts) >= 2:
+                    data['acc_georef'] = float(parts[1])
+    return data
+
+def extract_date_from_filename(filename):
+    """ Extract the date from the filename using regex. """
+    date_pattern = re.compile(r'(\d{4}-\d{2}-\d{2}-\d{2}-\d{2}-\d{2})')
+    match = date_pattern.search(filename)
+    if match:
+        return pd.to_datetime(match.group(1), format="%Y-%m-%d-%H-%M-%S")
+    return None
+
+def process_directory(directory, satellite_name):
+    """ Process each text file in the directory. """
+    records = []
+    for txt_file in glob.glob(os.path.join(directory, "*.txt")):
+        file_data = read_data_from_file(txt_file)
+        if 'filename' in file_data and 'acc_georef' in file_data:
+            date = extract_date_from_filename(file_data['filename'])
+            if date:
+                records.append({
+                    "satellite": satellite_name,
+                    "filename": file_data['filename'],
+                    "acc_georef": file_data['acc_georef'],
+                    "date": date
+                })
+    return records
+
+def save_data_to_csv(records, output_file):
+    """ Save the data to a CSV file. """
+    df = pd.DataFrame(records)
+    df.sort_values(by=["satellite", "date"], inplace=True)
+    df.to_csv(output_file, index=False)
+    print(f"Data saved to {output_file}")
+
+def plot_data(records):
+    """ Plot the date against acc_georef from the records, each satellite with different color. """
+    df = pd.DataFrame(records)
+    fig, ax = plt.subplots(figsize=(10, 6))
+    import matplotlib.colors as mcolors  # Correct import statement
+    colors = list(mcolors.TABLEAU_COLORS)  # Get a list of color names
+    for (label, group_df), color in zip(df.groupby('satellite'), colors):
+        group_df['month'] = group_df['date'].dt.month  # Extract month from date
+        ax.scatter(group_df["month"], group_df["acc_georef"], label=label, color=color, alpha=0.6)
+
+    ax.set_xlabel("Month")
+    ax.set_ylabel("acc_georef")
+    ax.set_title("acc_georef Over Months by Satellite")
+    ax.set_xticks(range(1, 13))  # Set x-ticks to be the months
+    ax.set_xticklabels(['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'])
+    plt.legend()
+    plt.tight_layout()
+    plt.show()
+
+
+#example : base_directory = r"C:\development\doodleverse\coastseg\CoastSeg\data\ID_1_datetime06-04-24__12_09_54"
+base_directory = ""
+satellites = [os.path.join(base_directory, d) for d in os.listdir(base_directory) if os.path.isdir(os.path.join(base_directory, d))]
+available_satellites = ['L5', 'L7', 'L8', 'L9']
+satellites = [satellite_dir for satellite_dir in satellites if os.path.basename(satellite_dir) in available_satellites]
+all_records = []
+for satellite_dir in satellites:
+    meta_dir = os.path.join(satellite_dir, "meta")
+    print(f"Processing {meta_dir}")
+    if os.path.exists(meta_dir):
+        records = process_directory(meta_dir,satellite_name=os.path.basename(satellite_dir))
+        all_records.extend(records)
+
+if all_records:
+    output_csv = "acc_georef_results.csv"
+    save_data_to_csv(all_records, output_csv)
+    plot_data(all_records)
+else:
+    print("No valid data found in the files.")