Spaces:
Sleeping
Sleeping
File size: 6,277 Bytes
e383754 4b38b88 67cd09b e383754 4b38b88 67cd09b e383754 4b38b88 824dd08 67cd09b 4b38b88 824dd08 67cd09b 4b38b88 67cd09b 824dd08 67cd09b 824dd08 4b38b88 824dd08 67cd09b 824dd08 67cd09b 824dd08 4b38b88 67cd09b 824dd08 67cd09b 824dd08 67cd09b 824dd08 67cd09b 4b38b88 824dd08 4b38b88 67cd09b 824dd08 67cd09b 824dd08 67cd09b 824dd08 67cd09b 4b38b88 67cd09b 824dd08 67cd09b 4b38b88 e383754 dbdc03c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 |
import os
import logging
import numpy as np
import rasterio
from rasterio.warp import transform_bounds
import cv2
from scipy import ndimage
def extract_features_from_geotiff(image_path, output_folder, feature_type="trees"):
"""Improved feature extraction with contour-based tree detection."""
try:
logging.info(f"Extracting {feature_type} from {image_path}")
with rasterio.open(image_path) as src:
logging.info(f"Image shape: {src.shape}, bands: {src.count}")
# Enhanced NDVI calculation
if src.count >= 3:
red = src.read(1).astype(float)
green = src.read(2).astype(float)
nir = src.read(4).astype(float) if src.count >= 4 else green
logging.info(f"Red band range: {red.min():.3f} to {red.max():.3f}")
logging.info(f"Green band range: {green.min():.3f} to {green.max():.3f}")
logging.info(f"NIR band range: {nir.min():.3f} to {nir.max():.3f}")
# Improved NDVI calculation
ndvi = np.divide(nir - red, nir + red + 1e-10)
# Adaptive thresholding based on image statistics
ndvi_mean = np.mean(ndvi)
ndvi_std = np.std(ndvi)
threshold = max(0.05, ndvi_mean + 0.1 * ndvi_std) # More lenient threshold
logging.info(f"NDVI mean: {ndvi_mean:.3f}, std: {ndvi_std:.3f}, threshold: {threshold:.3f}")
mask = ndvi > threshold
logging.info(f"Mask pixels above threshold: {np.sum(mask)} out of {mask.size}")
else:
band = src.read(1)
logging.info(f"Single band range: {band.min():.3f} to {band.max():.3f}")
# Adaptive threshold for single-band images
threshold = np.percentile(band, 40) # Reduced from 70 to be more lenient
logging.info(f"Single band threshold: {threshold:.3f}")
mask = band > threshold
logging.info(f"Mask pixels above threshold: {np.sum(mask)} out of {mask.size}")
# Get bounds
bounds = src.bounds
if src.crs:
west, south, east, north = transform_bounds(
src.crs, 'EPSG:4326',
bounds.left, bounds.bottom, bounds.right, bounds.top
)
else:
west, south, east, north = -74.1, 40.6, -73.9, 40.8
# Convert to uint8 for OpenCV processing
mask_uint8 = (mask * 255).astype(np.uint8)
logging.info(f"Mask uint8 range: {mask_uint8.min()} to {mask_uint8.max()}")
# Morphological operations to clean up the mask
kernel = np.ones((3, 3), np.uint8)
mask_cleaned = cv2.morphologyEx(mask_uint8, cv2.MORPH_CLOSE, kernel)
mask_cleaned = cv2.morphologyEx(mask_cleaned, cv2.MORPH_OPEN, kernel)
logging.info(f"After morphological ops: {np.sum(mask_cleaned > 0)} pixels")
# Find contours instead of using grid
contours, _ = cv2.findContours(mask_cleaned, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
logging.info(f"Found {len(contours)} contours")
# Filter contours by area and create features
features = []
height, width = mask.shape
min_area = (height * width) * 0.0001 # Minimum 0.01% of image area (reduced from 0.1%)
max_area = (height * width) * 0.2 # Maximum 20% of image area (increased from 10%)
logging.info(f"Area filter: min={min_area:.0f}, max={max_area:.0f}")
for i, contour in enumerate(contours):
area = cv2.contourArea(contour)
logging.info(f"Contour {i}: area={area:.0f}")
# Filter by area
if area < min_area or area > max_area:
logging.info(f"Contour {i}: filtered out (area {area:.0f} not in range)")
continue
# Simplify contour for better polygon shape
epsilon = 0.005 * cv2.arcLength(contour, True)
approx = cv2.approxPolyDP(contour, epsilon, True)
# Convert contour points to geographic coordinates
polygon_coords = []
for point in approx:
x, y = point[0]
# Convert pixel coordinates to geographic coordinates
x_ratio = x / width
y_ratio = y / height
lon = west + x_ratio * (east - west)
lat = north - y_ratio * (north - south)
polygon_coords.append([lon, lat])
# Close the polygon
if len(polygon_coords) > 2:
polygon_coords.append(polygon_coords[0])
# Calculate confidence based on area and shape
area_ratio = area / (height * width)
confidence = min(0.95, 0.5 + area_ratio * 10) # Adaptive confidence
feature = {
"type": "Feature",
"id": i,
"properties": {
"feature_type": feature_type,
"confidence": round(confidence, 2),
"area_pixels": int(area),
"area_ratio": round(area_ratio, 4)
},
"geometry": {
"type": "Polygon",
"coordinates": [polygon_coords]
}
}
features.append(feature)
logging.info(f"Contour {i}: added as feature with {len(polygon_coords)} points")
logging.info(f"Extracted {len(features)} tree features")
return {
"type": "FeatureCollection",
"features": features,
"feature_type": feature_type
}
except Exception as e:
logging.error(f"Error extracting features: {str(e)}")
return {"type": "FeatureCollection", "features": []} |